Health degree evaluation method, device and equipment of automatic train monitoring system and medium

Abstract

The invention relates to the technical field of rail transit, and provides a health degree evaluation method and device for an automatic train monitoring system, equipment and a medium. The method comprises the following steps: acquiring multi-dimensional data of each device in the automatic train monitoring system; performing health degree analysis based on the multi-dimensional data of each device to obtain an initial health degree of each device; obtaining a redundancy level, a device category and a position weight ratio of each device; adjusting the initial health degree of each device according to the redundancy level of each device to obtain the adjusted health degree of each device; determining the category health degree of each equipment category in the automatic train monitoring system according to the adjusted health degree of each equipment, the equipment category and the position weight proportion; and obtaining the category weight of each equipment category in the automatic train monitoring system, and determining the health degree of the automatic train monitoring system according to the category health degree and the category weight of each equipment category in the automatic train monitoring system.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and in particular to a method, apparatus, equipment and medium for evaluating the health of an automatic train monitoring system. Background Technology

[0002] The Automatic Train Supervision (ATS) system is a subsystem of the Automatic Train Control (ATC) system. Integrating data communication, computer, and signaling technologies, it is used in urban rail transit to manage and control trains and signaling equipment. The ATS system comprises various servers, workstations, and other equipment, and its operational stability is crucial to the overall safety of the line. Currently, health status assessments of ATS systems largely rely on single performance indicators or the qualitative experience of maintenance personnel. This biased and subjective approach can lead to untimely and inaccurate predictions of potential faults, thus affecting the reliability and availability of the ATS system. Therefore, improving the reliability of ATS health assessments has become an urgent technical challenge. Summary of the Invention

[0003] This application provides a method, apparatus, device, and medium for evaluating the health of an automatic train monitoring system, in order to solve the technical problem of how to improve the reliability of health evaluation of an automatic train monitoring system.

[0004] In a first aspect, embodiments of this application provide a health assessment method for an automatic train monitoring system, including: Acquire multi-dimensional data from each device in the automated train monitoring system. This multi-dimensional data is used to characterize the operating characteristics, historical performance, and environmental conditions of the corresponding devices. Health analysis was performed on each device based on multi-dimensional data to obtain the initial health status of each device. Obtain the redundancy level, device category, and location weight ratio for each device; The initial health of each device is adjusted according to its redundancy level to obtain the adjusted health of each device. Based on the adjusted health status of each device, device category, and location weight ratio, the category health status of each device category in the automatic train monitoring system is determined. Obtain the category weights of each equipment category in the automated train monitoring system, and determine the health status of the automated train monitoring system based on the category health status and category weights of each equipment category.

[0005] Secondly, embodiments of this application provide a health assessment device for an automatic train monitoring system, comprising: The first acquisition module is used to acquire multi-dimensional data of each device in the automatic train monitoring system. The multi-dimensional data is used to characterize the operating characteristics, historical performance and environmental status of the corresponding device. The analysis module is used to perform health analysis based on multi-dimensional data of each device to obtain the initial health status of each device. The second acquisition module is used to acquire the redundancy level, device category, and location weight ratio of each device. The adjustment module is used to adjust the initial health of each device according to the redundancy level of each device, so as to obtain the adjusted health of each device. The first determination module is used to determine the category health of each equipment category in the automatic train monitoring system based on the adjusted health of each equipment, equipment category, and location weight ratio. The second determining module is used to obtain the category weights of each equipment category in the automatic train monitoring system, and to determine the health of the automatic train monitoring system based on the category health and category weights of each equipment category in the automatic train monitoring system.

[0006] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the steps of the health evaluation method for the automatic train monitoring system of the first aspect.

[0007] Fourthly, embodiments of this application provide a non-transitory computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the steps of the health evaluation method for the automatic train monitoring system of the first aspect.

[0008] The health evaluation method, apparatus, equipment, and medium for an automated train monitoring system provided in this application first acquire multi-dimensional data of each device in the automated train monitoring system. This multi-dimensional data characterizes the corresponding device's operational characteristics, historical performance, and environmental status. Then, health analysis is performed on each device based on its multi-dimensional data to obtain its initial health status. Based on this, the redundancy level, device category, and position weight ratio of each device are acquired, and the initial health status of each device is adjusted according to its redundancy level to obtain its adjusted health status. Next, the category health status of each device category in the automated train monitoring system is determined based on its adjusted health status, device category, and position weight ratio. Finally, the category weight of each device category in the automated train monitoring system is acquired, and the health status of the automated train monitoring system is determined by combining the category health status and category weight of each device category. Thus, by integrating multi-dimensional data and introducing redundancy level, device category, position weight ratio, and category weight for hierarchical weighted calculation, the actual operating status of each device and even the entire automated train monitoring system can be comprehensively and objectively reflected, thereby effectively improving the reliability of the health evaluation of the automated train monitoring system. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating the health assessment method for an automatic train monitoring system provided in this application embodiment; Figure 2 This is a schematic diagram of the system architecture for health evaluation of the automatic train monitoring system provided in this application embodiment; Figure 3 This is a schematic diagram of the health evaluation device of the automatic train monitoring system provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0012] The Automatic Train Monitoring System (ATMS) is a subsystem of the Automatic Train Control System (ATS). It integrates data communication, computer, and signaling technologies and is used in urban rail transit to manage and control trains and signaling equipment. The ATS includes various servers, workstations, and other equipment, and its operational stability is crucial to the overall safety of the railway line.

[0013] In some related technologies, the health status of an automated train monitoring system is assessed by monitoring a single performance indicator of key equipment, such as processor utilization or network response time. However, this method limits its assessment to a localized operational situation and does not cover the overall status of all key equipment or other operating parameters. Therefore, the assessment results are one-sided and cannot comprehensively reflect the overall health status of the system.

[0014] Another related technology relies on qualitative assessments based on the experience of maintenance personnel. The assessment process depends on historical alarm data, equipment operation logs, and on-site inspections, combined with the maintenance personnel's personal knowledge. However, this method lacks standardized, quantifiable evaluation criteria, making the results susceptible to subjective influences, resulting in difficulties in reproducing the assessment process, and insufficient objectivity and consistency.

[0015] It is evident that the one-sidedness and subjectivity of the evaluation methods used in related technologies may lead to untimely and inaccurate predictions of potential faults, thereby affecting the reliability and availability of the automated train monitoring system. Therefore, improving the reliability of health assessments for automated train monitoring systems has become an urgent technical problem to be solved.

[0016] To address the aforementioned issues, the main solution provided in this application includes: first, acquiring multi-dimensional data of each device in the automatic train monitoring system, whereby the multi-dimensional data characterizes the corresponding device's operational features, historical performance, and environmental status; then, performing health analysis based on the multi-dimensional data of each device to obtain the initial health status of each device; next, acquiring the redundancy level, device category, and position weight ratio of each device, and adjusting the initial health status of each device according to its redundancy level to obtain the adjusted health status of each device; then, determining the category health status of each device category in the automatic train monitoring system based on the adjusted health status, device category, and position weight ratio of each device; finally, acquiring the category weight of each device category in the automatic train monitoring system, and combining the category health status and category weight of each device category to determine the overall health status of the automatic train monitoring system; thus, by integrating multi-dimensional data and introducing redundancy level, device category, position weight ratio, and category weight for hierarchical weighted calculation, the actual operational status of each device and even the entire automatic train monitoring system can be comprehensively and objectively reflected, thereby effectively improving the reliability of the health evaluation of the automatic train monitoring system.

[0017] The following provides a detailed description of the health evaluation method for the automatic train monitoring system provided in the embodiments of this application.

[0018] Please see Figure 1 , Figure 1 This is a flowchart illustrating a health assessment method for an automatic train monitoring system provided in an embodiment of this application. Figure 1 As shown, the method in this application embodiment may include the following steps S101-S106.

[0019] S101, acquire multi-dimensional data of each device in the automatic train monitoring system. The multi-dimensional data is used to characterize the operating characteristics, historical performance and environmental status of the corresponding device.

[0020] Specifically, to assess the health status of the Automated Train Monitoring System (ATMS), it is necessary to acquire multi-dimensional data from each device within the AMS. This multi-dimensional data characterizes the operational features, historical performance, and environmental conditions of the corresponding devices. The AMS refers to a distributed, real-time monitoring and control system integrating modern data communication, computer, network, and signaling technologies. The multi-dimensional data from each device within the AMS is a dataset used to assess the device's health status. Device operational features refer to parameters reflecting the device's working status during real-time operation, including but not limited to CPU utilization, memory utilization, hard disk utilization, and failover unit status of computer devices. Historical performance refers to the stability and reliability indicators exhibited by the device over past operating cycles, including but not limited to alarm levels, alarm frequency, mean time between failures (MTBF), and device history. Environmental conditions refer to the conditions of the physical environment in which the device is located, including but not limited to ambient temperature and humidity.

[0021] Regarding this step, in some possible implementations, communication with the relevant interfaces of each device in the automatic train monitoring system can be used to collect data characterizing the device's operating features, historical performance, and environmental conditions, and this collected data can be used as the multi-dimensional data. In some possible implementations, relevant data acquisition methods can be used to obtain relevant records of the operating features, historical performance, and environmental conditions of each device from the automatic train monitoring system, and then integrate them to obtain the multi-dimensional data.

[0022] S102 performs health analysis based on multi-dimensional data of each device to obtain the initial health status of each device.

[0023] Specifically, to assess the initial health of a single device, a health analysis needs to be performed on the multi-dimensional data of each device to obtain its initial health score. The health analysis refers to the process of calculating the multi-dimensional data of a device according to preset evaluation criteria to determine its health status; the initial health score of a device refers to the basic health score calculated based on its own multi-dimensional data.

[0024] Regarding this step, in some possible implementations, relevant evaluation functions can be used to process the multi-dimensional data to perform the health analysis, thereby obtaining the initial health of each device. In some possible implementations, relevant algorithms can be applied to the multi-dimensional data for mathematical operations to complete the health analysis and calculate the initial health of each device.

[0025] S103, obtain the redundancy level, equipment category and location weight ratio of each device.

[0026] Specifically, to incorporate the importance and deployment location weights of devices in the system into subsequent calculations to reflect the impact of device failures on the overall system, it is necessary to obtain the redundancy level, device category, and location weight ratio for each device. The redundancy level refers to the level categorized based on the degree of impact of device failure on operations, including core critical level, core general level, and non-security level. The device category refers to the category categorized based on device function and importance, including critical server category, non-critical server category, monitorable and controllable workstation category, monitorable but not controllable workstation category, and auxiliary device category. The location weight ratio refers to the weight ratio set according to the importance of the device's physical deployment location; for example, central devices, centralized station devices, critical station devices, and non-centralized station devices each correspond to different weight ratios.

[0027] Regarding this step, in some possible implementations, the redundancy level, device category, and location weight ratio of each device can be obtained by querying relevant configuration data. In other possible implementations, the redundancy level, device category, and location weight ratio of each device can be determined by reading relevant configuration files.

[0028] S104, adjust the initial health of each device according to the redundancy level of each device to obtain the adjusted health of each device.

[0029] Specifically, to reflect the impact of redundant configuration on the actual availability of equipment, the initial health of each device needs to be adjusted according to its redundancy level to obtain the adjusted health of each device. The adjusted health of a device refers to the health score obtained by adjusting the initial health based on the device's redundancy level and actual fault conditions, applying a preset deduction rule.

[0030] Regarding this step, in some possible implementations, relevant adjustment rules can be applied to process the initial health level. These adjustment rules are set based on the redundancy level to obtain the adjusted health level of each device. In some possible implementations, a relevant adjustment function can be used, taking the initial health level and the redundancy level as input, to calculate the adjusted health level of each device.

[0031] S105. Based on the adjusted health status of each device, device category, and location weight ratio, determine the category health status of each device category in the automatic train monitoring system.

[0032] Specifically, in order to aggregate the health status of individual devices to the device category level and assess the health status of different types of devices, it is necessary to determine the category health status of each device category in the automated train monitoring system based on the adjusted health status of each device, the device category, and the location weight ratio. Here, the device categories in the automated train monitoring system refer to the aforementioned critical server categories, non-critical server categories, etc.; the category health status of a device category refers to the score representing the health status of that device category, obtained by weighting the adjusted health status of all devices belonging to the same device category according to their location weight ratio.

[0033] Regarding this step, in some possible implementations, a relevant aggregation algorithm can be applied to the adjusted health score, which considers the device category and the location weight ratio to determine the category health score for each device category. In some possible implementations, a relevant weighted calculation can be performed, based on the adjusted health score and considering the device category and the location weight ratio, to obtain the category health score for each device category.

[0034] S106, obtain the category weights of each equipment category in the automatic train monitoring system, and determine the health status of the automatic train monitoring system based on the category health status and category weights of each equipment category in the automatic train monitoring system.

[0035] Specifically, in order to assess the health status of the entire automated train monitoring system, it is necessary to obtain the category weights of each equipment category in the automated train monitoring system, and determine the health status of the automated train monitoring system based on the category health and category weights of each equipment category. The health status of the automated train monitoring system refers to the evaluation value used to characterize the overall health status of the automated train monitoring system, calculated by weighted summation of the category health statuses of all equipment categories in the system.

[0036] Regarding this step, in some possible implementations, a related comprehensive calculation can be applied to the category health score, using the category weights as parameters to determine the health score of the automated train monitoring system. In some possible implementations, related mathematical operations can be performed, using the category health score and the category weights as inputs to calculate the health score of the automated train monitoring system.

[0037] In this embodiment, firstly, multi-dimensional data of each device in the automated train monitoring system is acquired. This multi-dimensional data characterizes the operating characteristics, historical performance, and environmental status of the corresponding devices. Then, health analysis is performed based on the multi-dimensional data of each device to obtain its initial health status. Next, the redundancy level, device category, and location weight ratio of each device are obtained. The initial health status of each device is then adjusted according to its redundancy level to obtain its adjusted health status. Following this, the category health status of each device category in the automated train monitoring system is determined based on its adjusted health status, device category, and location weight ratio. Finally, the category weight of each device category in the automated train monitoring system is obtained, and the overall health status of the automated train monitoring system is determined by combining the category health status and category weight. Thus, by integrating multi-dimensional data and introducing redundancy level, device category, location weight ratio, and category weight for hierarchical weighted calculation, the actual operating status of each device and even the entire automated train monitoring system can be comprehensively and objectively reflected, thereby effectively improving the reliability of the health evaluation of the automated train monitoring system.

[0038] In one embodiment, the step of "obtaining the redundancy level, device category, and location weight ratio of each device" can be further refined and may include the following steps: Based on the device identifier of each device and the first mapping relationship between the device identifier and the redundancy level, the redundancy level of each device is determined. The redundancy level includes at least one of the core critical level, core general level, or non-security level. Based on the device identifier of each device and the second mapping relationship between the device identifier and the device category, the device category of each device is determined. The device category includes at least one of the following: critical server category, non-critical server category, monitorable and controllable workstation category, monitorable but not controllable workstation category, and auxiliary device category. The location weight ratio of each device is determined based on its physical location and the third mapping relationship between physical location and location weight ratio.

[0039] Specifically, considering the importance and functional differences of each device in the automatic train monitoring system, as well as the impact of deployment location on the overall operation of the system, this embodiment proposes a method to obtain the redundancy level, device category, and location weight ratio of the device through device identification and physical location.

[0040] First, to accurately reflect the impact of equipment failures on system operation in subsequent calculations, it is necessary to determine the redundancy level of each device based on its device identifier and the primary mapping relationship between the device identifier and redundancy level. Redundancy level refers to the classification based on the degree of impact of equipment failure on operations; redundancy levels include at least one of the following: core critical level, core general level, or non-safety level. Core critical level refers to equipment where the failure would directly affect operations and cause significant losses; core general level refers to equipment where the failure would not directly affect operations, can be delayed, but requires vigilance; and non-safety level refers to equipment where the failure would not directly affect operations and is not urgently requiring repair.

[0041] Regarding this step, in some possible implementations, the device identifier of each device can be matched with the first mapping relationship by querying a preset configuration table or database, thereby determining the redundancy level of each device, and the determined redundancy level can be used as the redundancy level of each device.

[0042] Furthermore, to achieve categorized management of equipment categories in subsequent calculations, it is necessary to determine the equipment category of each device based on its equipment identifier and the second mapping relationship between the equipment identifier and the equipment category. The equipment category includes at least one of the following: critical server category, non-critical server category, monitorable and controllable workstation category, monitorable but not controllable workstation category, and auxiliary equipment category. Here, equipment category refers to the classification based on the equipment's function and importance.

[0043] Regarding this step, in some possible implementations, the device identifier of each device can be obtained by reading the device configuration file or calling the interface, and the device identifier of each device can be compared with the second mapping relationship to determine the device category of each device, and the determined device category can be used as the device category of each device.

[0044] Furthermore, to reflect the importance of the physical deployment location of the equipment in subsequent calculations, it is necessary to determine the location weight ratio of each device based on its physical location and the third mapping relationship between physical location and location weight ratio. Here, the location weight ratio refers to the weight ratio set according to the importance of the device's physical deployment location.

[0045] Regarding this step, in some possible implementations, the physical location information of each device can be obtained, and the physical location information of each device can be matched with the third mapping relationship to determine the location weight ratio of each device, and the determined location weight ratio can be used as the location weight ratio of each device.

[0046] In this embodiment, the redundancy level, equipment category, and location weight ratio of each device are obtained through device identification and physical location. This provides an accurate classification and weight basis for subsequent health calculations that combine the redundancy level, equipment category, and location weight ratio for hierarchical weighted calculations, thereby improving the reliability and accuracy of the health evaluation results of the automatic train monitoring system.

[0047] In one embodiment, the step "adjusting the initial health of each device according to the redundancy level of each device to obtain the adjusted health of each device" can be further refined and may include the following steps: Based on the redundancy level and fault status of each device, the initial health of each device is adjusted according to the preset deduction rules to obtain the adjusted health of each device. The deduction rules include: If the redundancy level of the equipment is at the core critical level and no failure occurs, the health of the equipment after adjustment shall not be lower than the preset first score. When the redundancy level of the equipment is at the core critical level and a single system failure occurs, the adjusted health of the equipment is between the preset second score and the preset first score. In the event of a fault that causes equipment malfunction, the adjusted health score of the equipment shall not exceed the preset third score. Among them, the preset first score is greater than the preset second score, and the preset second score is greater than the preset third score.

[0048] Specifically, considering the impact of equipment redundancy configuration on the actual availability of the system, and the different risks of different fault levels to system operation, this embodiment proposes a scheme to dynamically adjust the initial health by combining redundancy level and fault condition.

[0049] To accurately reflect the actual health status of equipment under a redundant architecture, the initial health of each device needs to be adjusted according to its redundancy level and fault condition, applying preset deduction rules to obtain the adjusted health status of each device. Here, the fault condition refers to whether the device has experienced a core critical fault, a single-system fault, or a functional failure; the deduction rules refer to the score adjustment logic set based on the device's redundancy level and fault type.

[0050] No equipment failure means that the equipment did not trigger any abnormal state affecting the operation of the system during the monitoring period; a single equipment failure means that a single equipment in the redundant configuration fails, but the system can still maintain operation through the redundancy mechanism; a failure that causes equipment to malfunction means that the equipment completely loses its intended function and cannot be restored to operation through redundancy switching.

[0051] Regarding this deduction rule, some possible implementation methods include: First, based on the redundancy level and fault status of each device, query a preset deduction threshold table, which contains preset first, second, and third scores corresponding to the core critical level; second, compare the initial health of the device with the scores in the threshold table. When the device meets the core critical level and has not experienced a fault, ensure that the adjusted health is not lower than the preset first score; when the device meets the core critical level but experiences a single-system fault, limit the adjusted health between the preset second score and the preset first score; when the device experiences a functional failure, force the adjusted health not to exceed the preset third score; finally, use the adjusted health score of each device as the adjusted health for subsequent category health calculations of the device.

[0052] In this embodiment, a differentiated adjustment of the initial health of equipment is achieved by matching the redundancy level of equipment with the deduction logic of fault type. This adjustment process sets upper or lower limits for the adjusted health of core critical level equipment that has not experienced a fault, core critical level equipment that has experienced a single-system fault, and equipment that has experienced a functional failure, based on preset first, second, and third scores corresponding to the equipment redundancy level. This constraint mechanism ensures that the adjusted health reflects the actual availability and risk level of the equipment under the redundancy architecture. Furthermore, the adjusted health score is used as the adjusted health, providing data input for the subsequent calculation steps of determining the category health of each equipment category in the automatic train monitoring system based on the adjusted health of each equipment, equipment category, and location weight ratio. This ensures that the health evaluation result of the entire automatic train monitoring system reflects the actual operating status of the system.

[0053] In one embodiment, the step "performing health analysis based on multi-dimensional data of each device to obtain the initial health of each device" can be further refined and may include the following steps: Obtain the base score for each device; Based on the hardware indicators that characterize the operating features of the equipment from multi-dimensional data, determine the hardware deduction value for each piece of equipment; Based on software metrics that characterize historical performance from multi-dimensional data, determine the software deduction value for each device; Based on environmental indicators that characterize the environmental state from multi-dimensional data, determine the environmental deduction value for each device; The initial health level of each device is determined based on its base score, hardware deduction score, software deduction score, and environmental deduction score.

[0054] Specifically, in order to comprehensively quantify the health status of the equipment, this embodiment proposes a scheme to calculate the initial health level by using a multi-dimensional index-based hierarchical deduction method.

[0055] First, we need to obtain the base score for each device. The base score refers to the preset maximum health score, which is used to characterize the health level of the device under ideal conditions.

[0056] Regarding this step, in some possible implementations, the base score can be set to a fixed value, such as 100 points, as the benchmark for subsequent deduction calculations.

[0057] Furthermore, based on hardware indicators characterizing device operation features from multi-dimensional data, a hardware deduction value is determined for each device. Hardware indicators refer to parameters reflecting the real-time operating status of the device, including but not limited to processor utilization, memory utilization, hard disk utilization, and failover unit status; the hardware deduction value is a deduction calculated based on the degree to which the hardware indicators deviate from preset standards.

[0058] Regarding this step, in some possible implementations, hardware parameters in multi-dimensional data can be parsed and compared with preset hardware indicator thresholds to generate corresponding hardware deduction values ​​based on the degree of deviation.

[0059] Furthermore, based on software metrics representing historical performance from multi-dimensional data, a software deduction value is determined for each device. Software metrics refer to parameters reflecting the historical operational stability of the device, including but not limited to alarm level, alarm frequency, mean time between failures (MTBF), and device history. The software deduction value is a deduction calculated based on the degree to which the software metrics deviate from preset standards.

[0060] Regarding this step, in some possible implementations, software operation records from multi-dimensional data can be statistically analyzed and compared with preset software indicator thresholds to generate corresponding software deduction values ​​based on the degree of deviation.

[0061] Furthermore, based on environmental indicators characterizing the environmental state from multi-dimensional data, environmental deduction values ​​are determined for each piece of equipment. Environmental indicators refer to parameters reflecting the physical environmental conditions of the equipment, including but not limited to ambient temperature and humidity; environmental deduction values ​​are calculated based on the degree to which environmental indicators deviate from preset standards.

[0062] Regarding this step, in some possible implementations, environmental parameters from multi-dimensional data can be collected and compared with preset environmental indicator thresholds to generate corresponding environmental deduction values ​​based on the degree of deviation.

[0063] Finally, the initial health of each device is determined based on its base score, hardware deduction score, software deduction score, and environmental deduction score.

[0064] Regarding this step, in some possible implementations, the initial health score of each device can be obtained by subtracting the hardware deduction, software deduction, and environmental deduction from the base score.

[0065] In this embodiment, by obtaining the basic score of each device, and based on hardware indicators representing device operating characteristics, software indicators representing historical performance, and environmental indicators representing environmental conditions from multi-dimensional data, the hardware deduction value, software deduction value, and environmental deduction value of each device are determined respectively. Then, based on the basic score, hardware deduction value, software deduction value, and environmental deduction value of each device, the initial health level of each device is determined, achieving a quantitative assessment of device health status. This assessment method integrates the real-time operating characteristics, historical performance, and environmental conditions of the device, avoiding the one-sidedness of single-indicator assessment. Simultaneously, through hierarchical deduction calculation, the initial health level can accurately reflect the actual health status of the device in different dimensions, providing an accurate data foundation for subsequent adjustments based on redundancy levels.

[0066] In one embodiment, the step "determining the hardware deduction value for each device based on hardware indicators characterizing device operation features from multi-dimensional data" can be further refined and may include the following steps: Processor utilization, memory utilization, and hard disk utilization, which characterize the device's operating features, are obtained from multi-dimensional data to determine the first hardware deduction value for each device. The reverse-cutting unit status, which characterizes the operating features of the equipment, is obtained from multi-dimensional data to determine the second hardware deduction value for each device; The hardware deduction value for each device is determined based on the first hardware deduction value and the second hardware deduction value for each device.

[0067] Specifically, the first step is to obtain processor utilization, memory utilization, and hard disk utilization—data that characterize device operation—from multi-dimensional data to determine the initial hardware deduction value for each device. Processor utilization refers to the percentage of time the processor is in a busy state per unit of time; memory utilization refers to the proportion of memory capacity used relative to the total memory capacity; hard disk utilization refers to the proportion of hard disk storage space used relative to the total hard disk storage space; and the initial hardware deduction value for each device is a deduction value calculated based on the degree to which processor utilization, memory utilization, and hard disk utilization deviate from preset standards.

[0068] Regarding this step, in some possible implementations, the processor utilization rate, memory utilization rate, and hard disk utilization rate can be compared with their respective preset utilization rate threshold ranges. Based on the comparison results, the corresponding deduction components can be determined, and then the deduction components can be accumulated or weighted to obtain the first hardware deduction value for each device.

[0069] Furthermore, the switchover unit status, which characterizes the operating features of the equipment, is obtained from multi-dimensional data to determine the second hardware deduction value for each device. Here, the switchover unit status refers to the operating status of the unit used for primary / standby switching in the redundant system; the second hardware deduction value for the device refers to the deduction value calculated based on the degree to which the switchover unit status deviates from a preset standard.

[0070] Regarding this step, in some possible implementations, it can be determined whether the switchover unit status in the multi-dimensional data is a fault state or a switchover failure state. If so, the second hardware deduction value is determined to be a preset deduction value; otherwise, the second hardware deduction value is determined to be zero, and this deduction value is used as the second hardware deduction value for each device.

[0071] Finally, based on the first and second hardware deduction values ​​of each device, the hardware deduction value of each device is determined.

[0072] Regarding this step, in some possible implementations, the first hardware deduction value and the second hardware deduction value of each device can be summed to obtain the hardware deduction value of each device.

[0073] In this embodiment, hardware indicators characterizing device operation are categorized, and a first hardware deduction value is determined based on processor utilization, memory utilization, and hard disk utilization. A second hardware deduction value is determined based on the switchover unit status. The final hardware deduction value is then determined by combining the first and second hardware deduction values, thus achieving differentiated processing for different types of hardware indicators. This processing method enables the hardware deduction value to comprehensively reflect the health status of the device in terms of computing resources, storage resources, and redundancy switching functions, improving the accuracy and comprehensiveness of the hardware deduction value and providing a basis for the accurate calculation of the initial health status.

[0074] In one embodiment, the step "determining the software deduction value for each device based on software metrics characterizing historical performance in multi-dimensional data" can be further refined and may include the following steps: The alarm level and alarm count, which characterize historical performance, are obtained from multi-dimensional data to determine the first software deduction value for each device. The mean time between failures (MTBF) and equipment history, which characterize historical performance, are obtained from multi-dimensional data to determine the second software deduction value for each device. The software deduction value for each device is determined based on the first software deduction value and the second software deduction value for each device.

[0075] Specifically, the first step is to obtain alarm levels and alarm counts representing historical performance from multi-dimensional data to determine the first software deduction value for each device. Here, alarm level refers to the classification based on the severity of the impact of alarm events on device operation; alarm count refers to the cumulative number of alarm events generated by the device within a preset statistical period; and the first software deduction value for the device is the deduction value calculated based on the degree to which the alarm level and alarm count deviate from preset standards.

[0076] Regarding this step, in some possible implementations, the alarm level can be matched with a preset level deduction rule to obtain a level deduction component, and the number of alarms can be matched with a preset number of alarms deduction rule to obtain a number of alarms deduction component. Then, the level deduction component and the number of alarms deduction component are combined and calculated to obtain the first software deduction value for each device.

[0077] Furthermore, the mean time between failures (MTBF) and equipment history, which characterize historical performance, are obtained from multi-dimensional data to determine the second software deduction value for each piece of equipment. Here, MTBF refers to the average operating time between two consecutive failures of the equipment; equipment history refers to information recording the equipment's commissioning duration, maintenance records, and lifecycle stages; and the second software deduction value is a deduction value calculated based on the degree to which the MTBF and equipment history deviate from preset standards.

[0078] Regarding this step, in some possible implementations, it is possible to determine whether the mean time between failures (MTBF) meets the preset qualification standard, and to determine the current life cycle stage of the equipment based on the equipment history. Then, according to the corresponding deduction rules, the MTBF deduction component and the history deduction component are generated respectively. Finally, the MTBF deduction component and the history deduction component are summarized and calculated, and the calculation result is used as the second software deduction value for each equipment.

[0079] Finally, based on the first and second software deduction values ​​of each device, the software deduction value of each device is determined.

[0080] Regarding this step, in some possible implementations, the first software deduction value and the second software deduction value of each device can be summed to obtain the software deduction value of each device.

[0081] In this embodiment, software indicators characterizing historical performance are categorized, and a first software deduction value is determined based on alarm level and alarm frequency. A second software deduction value is determined based on mean time between failures (MTBF) and equipment history. The final software deduction value is then determined by combining the first and second software deduction values, thus achieving differentiated processing for different types of software indicators. This processing method allows the software deduction value to comprehensively reflect the health status of the equipment in terms of immediate alarm conditions and long-term operational stability, improving the accuracy and comprehensiveness of the software deduction value and providing a basis for the accurate calculation of the initial health status.

[0082] In one embodiment, the step of "determining the environmental deduction value of each device based on environmental indicators characterizing the environmental state from multi-dimensional data" can be further refined and may include the following steps: Ambient temperature and humidity, which characterize the environmental state, are obtained from multi-dimensional data to determine the environmental deduction value for each device.

[0083] Specifically, the first step is to obtain environmental temperature and humidity data, which characterize the environmental state, from multi-dimensional data to determine the environmental deduction value for each device. Environmental temperature refers to the temperature value of the physical space where the device is located; environmental humidity refers to the relative humidity value of the physical space where the device is located.

[0084] Regarding this step, in some possible implementations, the ambient temperature and ambient humidity can be compared with their respective preset environmental threshold ranges. If either indicator exceeds the preset environmental threshold range, the corresponding deduction value is determined according to the preset deduction rules, and the obtained deduction value is used as the environmental deduction value for each device. If both are within the preset environmental threshold range, the environmental deduction value is determined to be zero.

[0085] In this embodiment, by acquiring ambient temperature and humidity, which characterize the environmental state, and comparing them with preset thresholds to determine the environmental deduction value, a quantitative assessment of the equipment operating environment factors is achieved. This assessment method incorporates physical environmental conditions into the health calculation, enabling the environmental deduction value to reflect the potential impact of environmental factors on equipment operation, thereby providing a more comprehensive basis for subsequent initial health calculations.

[0086] In one embodiment, the step of "determining the category health of each device category in the automatic train monitoring system based on the adjusted health of each device, device category, and location weight ratio" can be further refined and may include the following steps: Based on the equipment category of each device, the devices belonging to the same equipment category in the automatic train monitoring system are grouped to obtain at least one equipment group; The adjusted health score of each device in each device group is multiplied by the position weight ratio to obtain the weighted health score of each device in each device group. Based on the weighted health scores of each device within each device group, the category health of the corresponding device category for each device group is determined.

[0087] Specifically, considering the diversity of equipment categories in the automatic train monitoring system and the varying importance of the physical deployment location of the equipment to the overall system operation, this embodiment proposes to assess the health status of various types of equipment by grouping them by equipment category and combining the weight ratio of location.

[0088] First, based on the equipment category of each device, the devices in the automatic train monitoring system belonging to the same category need to be grouped to obtain at least one equipment group. Here, an equipment group refers to a set of devices of the same category, which is used for subsequent unified health calculations.

[0089] Regarding this step, in some possible implementations, it can be accomplished by traversing all devices in the automatic train monitoring system, classifying each device according to its equipment category, and grouping devices of the same category into the same equipment group.

[0090] Furthermore, the adjusted health score of each device within each device group is multiplied by its location weight ratio to obtain a weighted health score for each device within each device group. The weighted health score is a score calculated by weighting the adjusted health score with the importance of the device's physical deployment location, reflecting the device's contribution to health at a specific location.

[0091] Regarding this step, in some possible implementations, for each device group, the position weight ratio of each device in the group can be obtained sequentially, and the adjusted health of each device can be multiplied by the corresponding position weight ratio. The calculation result can be used as the weighted health score of each device.

[0092] Ultimately, to achieve a quantitative assessment of the health status of equipment categories, it is necessary to determine the category health score of each equipment category within each equipment group based on the weighted health scores of all equipment within that group. The category health score of an equipment group refers to the score calculated by aggregating the weighted health scores of all equipment within that group, and is used to characterize the overall health level of that equipment category.

[0093] Regarding this step, in some possible implementations, the weighted health scores of all devices within each device group can be summed or averaged, and the calculation result can be used as the category health score of the corresponding device category for the device group, which can then be used for the subsequent health assessment of the automatic train monitoring system.

[0094] In this embodiment, devices are grouped according to their device categories, achieving unified management of devices of the same type. By multiplying the adjusted health score of each device within each group by its location weight, a weighted health score is obtained, reflecting the difference in the contribution of the physical deployment location to the device's health score. Based on the weighted health scores of each device within each group, the category health score of the corresponding device category for each group is determined, thus aggregating the health status of all devices within the same category. Therefore, this embodiment enables the category health score to comprehensively reflect the actual health level of that category of devices in the system, providing a reliable basis for subsequently determining the overall health of the automated train monitoring system based on the category health score and category weight of each device category.

[0095] In one embodiment, the step of "determining the health of the automatic train monitoring system based on the category health and category weight of each equipment category in the automatic train monitoring system" can be further refined and may include the following steps: The weighted sum is obtained by weighting and summing the health status and weight of each equipment category in the automatic train monitoring system. The health status of the automatic train monitoring system is determined based on the weighted summation result.

[0096] Specifically, considering the impact of the category health and category weight of each equipment category in the automatic train monitoring system on the overall system health, as well as the differences in importance of different equipment categories in the system, this embodiment proposes a scheme to comprehensively calculate the system health by weighted summation.

[0097] First, a weighted sum needs to be calculated based on the health status and weight of each equipment category in the automatic train monitoring system to obtain the weighted sum result.

[0098] Regarding this step, in some possible implementations, the category health score and category weight of each equipment category in the automatic train monitoring system can be obtained, and the category health score of each category can be multiplied by the corresponding category weight to obtain the weighted health score of each category. Then, the weighted health scores of each category can be summed to obtain the weighted sum result.

[0099] Furthermore, the health status of the automatic train monitoring system is determined based on the weighted summation result.

[0100] Regarding this step, in some possible implementations, the weighted summation result can be directly used as the health status of the automatic train monitoring system; or, the weighted summation result can be normalized and the normalized result can be used as the health status of the automatic train monitoring system.

[0101] In this embodiment, the weighted sum of the category health scores of each equipment category in the automated train monitoring system and their corresponding category weights is obtained. This weighted sum is then used to determine the overall health score of the automated train monitoring system, thus achieving a quantitative assessment of the system's overall health status. This processing method combines the category health scores and category weights of each equipment category, allowing for a differentiated representation of the contribution of different equipment categories to the overall system health, thereby reflecting the actual health status of each equipment category within the system.

[0102] To facilitate understanding of the above embodiments, the method provided in this application will be described in detail below with reference to a specific embodiment. In this embodiment, the health evaluation of an automatic train monitoring system will be used as an example to specifically illustrate how to calculate the health score of the automatic train monitoring system through multi-dimensional data, hierarchical weights, and preset rules.

[0103] First, an equipment classification and weighting system is established for the evaluation framework. In this embodiment, the equipment in the ATS system is divided into categories based on importance: critical servers, non-critical servers, monitorable and controllable workstations, monitorable but not controllable workstations, and auxiliary equipment. A category weight is assigned to each equipment category. Simultaneously, a redundancy level is assigned to each device based on the impact of its failure on operations, and a location weight ratio is assigned based on its physical deployment location.

[0104] The equipment categories and their weights can be configured as follows: critical servers (50%), non-critical servers (10%), monitorable and controllable workstations (20%), workstations that are only monitored but not controlled (10%), and auxiliary equipment (10%).

[0105] The specific equipment classifications, category weights, and redundancy levels for each piece of equipment are shown in Table 1 below: Table 1 Equipment Classification Table Subsystem Classification Equipment Name Redundancy level ATS Critical servers (50%) Application Server A core and key ATS Critical servers (50%) Application Server B core and key ATS Critical servers (50%) Communication front-end unit A core and key ATS Critical servers (50%) Communication front-end B core and key ATS Critical servers (50%) Database server A core and key ATS Critical servers (50%) Database server B core and key ATS Critical servers (50%) Disk Array core and key ATS Critical servers (50%) Gateway computer A core and key ATS Critical servers (50%) Gateway Computer B core and key ATS Critical servers (50%) Interface server core and key ATS Critical servers (50%) System Interface Workstation A Core general ATS Critical servers (50%) System Interface Workstation B Core general ATS Non-critical servers (10%) COCC Interface Machine A Unsafe ATS Non-critical servers (10%) COCC Interface Machine B Unsafe ATS Non-critical servers (10%) Large screen interface server Unsafe ATS Monitorable and controllable workstations (20%) Administrator workstation Core general ATS Monitorable and controllable workstations (20%) Dispatch Workstation 1 Core general ATS Monitorable and controllable workstations (20%) Dispatch Workstation 2 Core general ATS Monitorable and controllable workstations (20%) General Dispatch Workstation Core general ATS Monitorable and controllable workstations (20%) Station extension Core general ATS Monitorable and controllable workstations (20%) Local workstation Core general ATS Workstations that are monitored but not controlled (10%) ATS Monitoring Workstation Unsafe ATS Ancillary equipment (10%) Departure timer Unsafe Redundancy levels are categorized as follows: Core Critical Level, indicating that the fault will directly impact operations and cause significant losses; Core General Level, indicating that the fault will not directly impact operations and can be addressed later, but vigilance is required; and Non-Safe Level, indicating that the fault will not directly impact operations and is not urgent for repair. As shown in Table 1, application servers and database servers are designated as Core Critical Level; system interface workstations and dispatch workstations are designated as Core General Level; and COCC interface machines and departure timers are designated as Non-Safe Level.

[0106] The location weight ratio can be set according to the equipment deployment location, for example: central equipment (50%), centralized station equipment (30%), key station equipment (15%), and non-centralized station equipment (5%).

[0107] Secondly, the multi-dimensional evaluation indicators and the calculation rules for the initial health status are defined. The initial health status of the equipment is calculated by subtracting various deduction values ​​from a base score (e.g., 100 points). The deduction values ​​are determined based on hardware indicators characterizing the equipment's operational characteristics, software indicators characterizing historical performance, and environmental indicators characterizing the environmental state from the multi-dimensional data.

[0108] Specifically, the deductions for hardware metrics include: Processor utilization rate: The deduction criteria are as follows: less than 30%, no points deducted; greater than 30% but less than 50%, deduct 10 points; greater than 50% but less than 70%, deduct 20 points; greater than 70% but less than 90%, deduct 30 points; greater than 90%, deduct 100 points. Its evaluation function can be expressed as: ; Memory usage rate: The deduction criteria are as follows: less than 30%, no points deducted; greater than 30% but less than 40%, deduct 10 points; greater than 40% but less than 60%, deduct 20 points; greater than 60% but less than 80%, deduct 30 points; greater than 80%, deduct 100 points. Its evaluation function can be expressed as: ; Hard drive usage rate: The deduction criteria are as follows: less than 60%, no points deducted; greater than 60% but less than 70%, deduct 10 points; greater than 70% but less than 80%, deduct 20 points; greater than 80%, deduct 100 points. Its evaluation function can be expressed as: ; Switching unit status: The deduction standard is that if a fault or switching failure occurs, 100 points will be deducted.

[0109] Specifically, the deductions for software metrics include: Alarm levels: The deduction standard is as follows: Level 1 alarm deducts 100 points; Level 2 alarm deducts 30 points; Level 3 alarm deducts 15 points.

[0110] Number of alarms: The deduction standard is as follows: 1 alarm, no points are deducted; 2 alarms, 30 points are deducted; 3 or more alarms, 100 points are deducted.

[0111] Mean Time Between Failures (MTBF): The deduction standard is 10 points for a non-compliant MTBF, and no points are deducted for a compliant one.

[0112] Equipment history: The deduction standard is as follows: no points are deducted if the equipment lifespan is below 50%; 5 points are deducted if the equipment lifespan is below 80%; and 10 points are deducted if the equipment lifespan is above 80%.

[0113] Specifically, the environmental indicators that result in point deductions include: Environment (temperature, humidity, etc.): The deduction standard is 5 points for unqualified environment, and no points are deducted for qualified environment.

[0114] To more clearly illustrate the specific evaluation criteria for each indicator, please refer to Table 2 below: Table 2 Evaluation Tables for Various Indicators Classification equipment standard rating level Hardware environment processor utilization Less than 30% excellent Hardware environment processor utilization Greater than 30% and less than 50% good Hardware environment processor utilization Greater than 50% and less than 70% generally Hardware environment processor utilization Greater than 70% and less than 90% nervous Hardware environment processor utilization Greater than 90% serious Hardware environment Memory usage Less than 30% excellent Hardware environment Memory usage Greater than 30% and less than 40% good Hardware environment Memory usage Greater than 40% and less than 60% generally Hardware environment Memory usage Greater than 60% and less than 80% nervous Hardware environment Memory usage Greater than 80% serious Hardware environment Hard drive usage Less than 50% excellent Hardware environment Hard drive usage Greater than 50% and less than 60% good Hardware environment Hard drive usage Greater than 60% and less than 70% generally Hardware environment Hard drive usage Greater than 70% and less than 80% nervous Hardware environment Hard drive usage Greater than 80% serious Hardware environment Backcut unit Fault / Reverse Cut Failure serious Software status Alarm Level Level 1 alarm serious Software status Alarm Level Level 2 alarm nervous Software status Alarm Level Level 3 alarm generally Software status Number of alarms One alarm generally Software status Number of alarms Two alarms nervous Software status Number of alarms More than 2 alarms serious Software status MTBF qualified / Software status MTBF Unqualified / Finally, a hierarchical weighted calculation is performed to determine the system's health.

[0115] Calculate the adjusted health score of the equipment. After calculating the initial health score of each piece of equipment according to the above rules, adjust it based on its redundancy level and fault status, applying preset deduction rules. For example, if the equipment is at the "core critical" level and no faults have occurred, its adjusted health score should not be lower than 70 points; if a single-system fault occurs, its adjusted health score should be between 60 and 70 points; if a functional failure occurs, its adjusted health score should not exceed 59 points.

[0116] Calculate the category health score for each device category. Group devices belonging to the same category (e.g., all "critical servers"). Multiply the adjusted health score of each device in the group by its location weight to obtain a weighted health score for each device. Sum the weighted health scores of all devices in the group to obtain the category health score for that device category.

[0117] Calculate the health score of the automated train monitoring system. Multiply the health score of each equipment category by its category weight to obtain a weighted health score for each category. Sum the weighted health scores of all categories to obtain the final health score of the automated train monitoring system.

[0118] In this embodiment, by specifically defining equipment classification, weighting system, evaluation indicators and deduction rules, and elaborating on the complete calculation process from the health status of individual devices to the health status of the system, the health status evaluation process is made more standardized, transparent and accurate, providing maintenance personnel with quantitative decision-making basis, thereby effectively improving the accuracy and efficiency of fault diagnosis.

[0119] To facilitate understanding of the present application, this embodiment provides a complete implementation process for a health evaluation method for an automatic train monitoring system based on hierarchical aggregation and multidimensional weight allocation. This process revolves around... Figure 2 The system architecture shown is expanded as follows.

[0120] First, the station-side equipment fault monitoring unit and the central-side equipment fault monitoring unit are activated in parallel to collect data from equipment deployed at different physical locations within the automatic train monitoring system. The station-side equipment fault monitoring unit refers to a module used to monitor the real-time operating status of all managed equipment within the rail transit station area. The central-side equipment fault monitoring unit refers to a module used to monitor the real-time operating status of all core equipment within the control center area. Both collect multi-dimensional data from each device. This multi-dimensional data refers to the raw data set characterizing the corresponding device's operating characteristics at the current moment, its historical cumulative performance, and its environmental state. The collected results are referred to as the data acquisition results.

[0121] Next, the data acquisition results are sent to two parallel fault evaluation and deduction modules associated with unique device identifiers. These modules are central processing modules that match and invoke corresponding fault evaluation standards and deduction strategies based on the unique device identifier. Within these modules, the system parses the data acquisition results and extracts the device identifier for each device. The device identifier is a code or name used to uniquely identify a device in the automatic train monitoring system. Simultaneously, based on alarm events and status changes generated by the device during the monitoring period, first fault event data is generated. This first fault event data serves as a component representing the real-time fault status of the device in the data acquisition results. Subsequently, this module queries preset first and second mapping relationships based on the device identifier to determine the redundancy level and device category for each device. This information, along with the first fault event data, forms the initial data package for device evaluation, used for subsequent single-device health assessments.

[0122] Subsequently, the initial device assessment data packet is input to the single-device health assessment unit (including redundancy determination function). The single-device health assessment unit (including redundancy determination function) is a functional unit responsible for independently analyzing the health of a single device and preliminarily determining its redundant configuration status. This unit first obtains the base score for each device, which refers to a preset maximum health evaluation score, such as 100 points. Then, the unit decomposes the initial device assessment data packet, extracting hardware indicators characterizing device operating characteristics, software indicators characterizing historical performance, and environmental indicators characterizing environmental status. Specifically, the unit generates first hardware assessment data by parsing the hardware parameters in the initial device assessment data packet. This first hardware assessment data serves as the hardware indicators characterizing the device operating characteristics, including processor utilization, memory utilization, hard disk utilization, and failover unit status. Simultaneously, the unit generates first software assessment data by statistically analyzing the software operation records in the initial device assessment data packet. This first software assessment data serves as the software indicators characterizing historical performance, including alarm level, alarm frequency, mean time between failures (MTBF), and device history. Furthermore, this unit generates first environmental assessment data by collecting environmental parameters from the initial data packet of the device assessment. This first environmental assessment data serves as an environmental indicator characterizing the environmental state, and the environmental indicator includes ambient temperature and humidity. Based on this, the unit calculates corresponding hardware deduction values, software deduction values, and environmental deduction values ​​based on the first hardware assessment data, first software assessment data, and first environmental assessment data, respectively, according to preset deduction thresholds. Finally, the unit subtracts the hardware deduction value, software deduction value, and environmental deduction value from the base score to obtain the initial health level of each device.

[0123] Then, the obtained initial health score, along with the redundancy level information contained in the initial device assessment data packet, is passed to the redundancy level-driven deduction execution module. This redundancy level-driven deduction execution module refers to the module that performs a secondary adjustment of the initial health score based on the device's redundancy level and real-time fault conditions, applying preset deduction rules. This module retrieves the first fault event data from the initial device assessment data packet to determine whether the device has experienced a core critical component failure, a single-system failure, or a functional failure. Combining the known redundancy level (core critical level, core general level, or non-safety level), this module applies preset deduction rules to adjust the initial health score: for example, if the device's redundancy level is core critical level and no failure has occurred, the adjusted health score is ensured to be no lower than a preset first score (e.g., 70 points); if a single-system failure occurs, the adjusted health score is limited to between a preset second score (e.g., 60 points) and the preset first score; if a failure occurs that causes device functional failure, the adjusted health score is forced to be no higher than a preset third score (e.g., 59 points). The adjusted health scores of each device will serve as the basis for subsequent aggregation calculations.

[0124] Next, the adjusted health scores of each device originating from the station side and the center side are aggregated into a device-class health score aggregation unit (no redundancy scenario). This unit (no redundancy scenario) aggregates the health scores of individual devices according to their device categories to assess the overall health level of each type of device under the assumption of no redundancy weighting. Based on the device category of each device, this unit groups devices belonging to the same device category within the automated train monitoring system, obtaining at least one device group. Each device group refers to a set of devices belonging to the same device category. Subsequently, this unit aggregates the adjusted health scores of all devices within each device group and calculates the average score for each group. This average score is used as the median category health score for the corresponding device category, and each device category and its corresponding median category health score form a category health score dataset.

[0125] The category health dataset is sent to a multi-type functional device group (including server group, interface group, workstation group, etc.). This multi-type functional device group (including server group, interface group, workstation group, etc.) refers to a module that logically divides and organizes the category health dataset according to the functional role of the devices in the system. Based on the functional descriptions of the devices, this module divides the device category health dataset into logical groups such as server group, interface group, and workstation group, forming grouped health data.

[0126] The grouped health data then enters the equipment group weight allocation and health calculation unit (50% for central stations, 30% for centralized stations, 15% for critical stations, and 5% for non-central stations). This unit (50% for central stations, 30% for centralized stations, 15% for critical stations, and 5% for non-central stations) refers to the unit that performs weighted calculations on the grouped health data based on the weight ratio of the equipment's physical deployment location. This unit reads the physical location of each device and determines the location weight ratio for each device based on a third mapping relationship between physical location and location weight ratio. The location weight ratio includes 50% for central stations, 30% for centralized stations, 15% for critical stations, and 5% for non-central stations. This unit multiplies the adjusted health score of each device by its location weight ratio to obtain the location-weighted health score for each device, and then summarizes the results into location-weighted health score data.

[0127] The location-weighted health score data is transmitted to the equipment category classification unit (including critical equipment, non-critical equipment, monitorable and controllable workstations, monitor-only workstations, and auxiliary equipment). This equipment category classification unit (including critical equipment, non-critical equipment, monitorable and controllable workstations, monitor-only workstations, and auxiliary equipment) refers to a unit that categorizes equipment into preset major categories based on its importance and function. This unit precisely classifies all equipment into five major categories—critical equipment, non-critical equipment, monitorable and controllable workstations, monitor-only workstations, and auxiliary equipment—based on their equipment category, and aggregates the location-weighted health scores of all equipment within each major category.

[0128] The aggregated data is then processed by the category-based health weight aggregation unit (within the same category, the average weight is applied). This category-based health weight aggregation unit (within the same category, the average weight is applied) is the unit that averages the device health scores within each category. This unit calculates the arithmetic mean of the location-weighted health scores of all devices within each category to obtain the average health score for each category, and uses this average health score as the provisional category health score.

[0129] Finally, the temporary category health score is input into the category health score weight configuration unit (configured as 50%, 10%, 20%, 10%, 10%). This unit (configured as 50%, 10%, 20%, 10%, 10%) is the unit that loads and applies the final category weights to calculate the overall system health score. This unit obtains the category weights for each equipment category in the Automated Train Monitoring System (ATS), configured as follows: critical equipment 50%, non-critical equipment 10%, monitorable and controllable workstations 20%, monitor-only workstations 10%, and auxiliary equipment 10%. This unit performs a weighted summation of the temporary category health score for each category with its corresponding category weight, obtaining the final weighted summation result, which is then determined as the overall health score of the ATS. The target ATS system health assessment object refers to the ATS itself, which is used as the overall health assessment target.

[0130] The following will combine Figure 3 The health evaluation device 800 for an automatic train monitoring system provided in this application embodiment will be described in detail. The health evaluation device 800 for the automatic train monitoring system can be referred to in correspondence with the health evaluation method for the automatic train monitoring system described above. Specifically, the health evaluation device 800 for the automatic train monitoring system may include a first acquisition module 810, an analysis module 820, a second acquisition module 830, an adjustment module 840, a first determination module 850, and a second determination module 860, as detailed below: The first acquisition module 810 is used to acquire multi-dimensional data of each device in the automatic train monitoring system. The multi-dimensional data is used to characterize the operating characteristics, historical performance and environmental status of the corresponding device. Analysis module 820 is used to perform health analysis based on multi-dimensional data of each device to obtain the initial health of each device. The second acquisition module 830 is used to acquire the redundancy level, equipment category and location weight ratio of each device. The adjustment module 840 is used to adjust the initial health of each device according to the redundancy level of each device, so as to obtain the adjusted health of each device. The first determining module 850 is used to determine the category health of each equipment category in the automatic train monitoring system based on the adjusted health of each equipment, equipment category, and position weight ratio. The second determining module 860 is used to obtain the category weights of each equipment category in the automatic train monitoring system, and to determine the health of the automatic train monitoring system based on the category health and category weights of each equipment category in the automatic train monitoring system.

[0131] Optionally, in some embodiments of this application, the specific execution steps of the health assessment device 800, the first acquisition module 810, the analysis module 820, the second acquisition module 830, the adjustment module 840, the first determination module 850, and the second determination module 860 can be referred to the above method embodiments, and will not be repeated here.

[0132] The effects achievable in this embodiment can be found in the relevant embodiments of the health evaluation method for the automatic train monitoring system described above, and will not be repeated here.

[0133] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include: a processor 1301, a communication interface 1302, a memory 1303, and a communication bus 1304, wherein the processor 1301, the communication interface 1302, and the memory 1303 communicate with each other via the communication bus 1304. The processor 1301 can call a computer program in the memory 1303 to execute the steps of the health evaluation method of the automatic train monitoring system, such as including: Acquire multi-dimensional data from each device in the automated train monitoring system. This multi-dimensional data is used to characterize the operating characteristics, historical performance, and environmental conditions of the corresponding devices. Health analysis was performed on each device based on multi-dimensional data to obtain the initial health status of each device. Obtain the redundancy level, device category, and location weight ratio for each device; The initial health of each device is adjusted according to its redundancy level to obtain the adjusted health of each device. Based on the adjusted health status of each device, device category, and location weight ratio, the category health status of each device category in the automatic train monitoring system is determined. Obtain the category weights of each equipment category in the automated train monitoring system, and determine the health status of the automated train monitoring system based on the category health status and category weights of each equipment category.

[0134] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0135] On the other hand, embodiments of this application also provide a non-transitory computer-readable storage medium storing a computer program. The computer program is used to cause a processor to execute the steps of the methods provided in the above embodiments, including, for example: Acquire multi-dimensional data from each device in the automated train monitoring system. This multi-dimensional data is used to characterize the operating characteristics, historical performance, and environmental conditions of the corresponding devices. Health analysis was performed on each device based on multi-dimensional data to obtain the initial health status of each device. Obtain the redundancy level, device category, and location weight ratio for each device; The initial health of each device is adjusted according to its redundancy level to obtain the adjusted health of each device. Based on the adjusted health status of each device, device category, and location weight ratio, the category health status of each device category in the automatic train monitoring system is determined. Obtain the category weights of each equipment category in the automated train monitoring system, and determine the health status of the automated train monitoring system based on the category health status and category weights of each equipment category.

[0136] Non-transitory computer-readable storage media can be any available medium or data storage device that can be accessed by a processor, including but not limited to magnetic storage (e.g., floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (e.g., ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0137] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0138] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for evaluating the health of an automatic train supervision system, characterized in that The method comprises: obtaining multi-dimensional data of each device in an automatic train supervision system, the multi-dimensional data being used to represent running characteristics, historical performance and environmental state of the corresponding device; respectively performing health degree analysis based on the multi-dimensional data of each device to obtain initial health degrees of each device; obtaining redundancy levels, device categories and location weight proportions of each device; respectively adjusting the initial health degrees of each device according to the redundancy levels of each device to obtain adjusted health degrees of each device; determining category health degrees of each device category in the automatic train supervision system according to the adjusted health degrees of each device, the device categories and the location weight proportions; obtaining category weights of each device category in the automatic train supervision system, and determining a health degree of the automatic train supervision system according to the category health degrees of each device category and the category weights.

2. The method of claim 1, wherein, The method comprises: determining the redundancy levels of each device according to device identifiers of each device and a first mapping relationship between device identifiers and redundancy levels, wherein the redundancy levels comprise at least one of a core key level, a core general level or a non-safety level; determining the device categories of each device according to the device identifiers of each device and a second mapping relationship between device identifiers and device categories, wherein the device categories comprise at least one of a key server category, a non-key server category, a monitorable and controllable workstation category, a monitor-only workstation category and an auxiliary device category; determining the location weight proportions of each device according to physical positions of each device and a third mapping relationship between physical positions and location weight proportions.

3. The method of claim 2, wherein, The method comprises: respectively adjusting the initial health degrees of each device according to the redundancy levels and failure conditions of each device by using a preset deduction rule to obtain the adjusted health degrees of each device; The deduction rule comprises: in the case that the redundancy level of the device is the core key level and no failure occurs, the adjusted health degree of the device is not lower than a preset first score value; in the case that the redundancy level of the device is the core key level and a single-system failure occurs, the adjusted health degree of the device is between a preset second score value and the preset first score value; in the case that a failure causing device function failure occurs, the adjusted health degree of the device is not higher than a preset third score value; wherein the preset first score value is greater than the preset second score value, and the preset second score value is greater than the preset third score value.

4. The method of claim 1, wherein, The method comprises: obtaining a basic score value of each device; determining a hardware deduction score value of each device based on a hardware index representing the running characteristics of the device in the multi-dimensional data; determine a software deduction value of each of the devices based on the software indicators representing the historical performance in the multi-dimensional data; determine an environment deduction value of each of the devices based on the environment indicators representing the environment state in the multi-dimensional data; determine an initial health degree of each of the devices according to the basic score value, the hardware deduction value, the software deduction value and the environment deduction value of each of the devices.

5. The method of claim 4, wherein, The determining of the hardware deduction value of each of the devices based on the hardware indicators representing the device running features in the multi-dimensional data comprises: obtaining processor occupancy, memory occupancy and hard disk usage representing the device running features from the multi-dimensional data to determine a first hardware deduction value of each of the devices; obtaining a state of a back-off unit representing the device running features from the multi-dimensional data to determine a second hardware deduction value of each of the devices; determining the hardware deduction value of each of the devices based on the first hardware deduction value and the second hardware deduction value of each of the devices; The determining of the software deduction value of each of the devices based on the software indicators representing the historical performance in the multi-dimensional data comprises: obtaining alarm level and alarm frequency representing the historical performance from the multi-dimensional data to determine a first software deduction value of each of the devices; obtaining mean time between failures and device history representing the historical performance from the multi-dimensional data to determine a second software deduction value of each of the devices; determining the software deduction value of each of the devices based on the first software deduction value and the second software deduction value of each of the devices; The determining of the environment deduction value of each of the devices based on the environment indicators representing the environment state in the multi-dimensional data comprises: obtaining environment temperature and environment humidity representing the environment state from the multi-dimensional data to determine the environment deduction value of each of the devices.

6. The method of claim 1, wherein, The determining of the category health degree of each device category in the automatic train supervision system according to the adjusted health degree, the device category and the location weight proportion of each of the devices comprises: grouping the devices belonging to the same device category in the automatic train supervision system according to the device category of each of the devices to obtain at least one device group; performing multiplication operation on the adjusted health degree and the location weight proportion of each of the devices in each of the device groups to obtain a weighted health score of each of the devices in each of the device groups; determining the category health degree of the corresponding device category of each of the device groups according to the weighted health score of each of the devices in each of the device groups.

7. The method of claim 1, wherein, The determining of the health degree of the automatic train supervision system according to the category health degree and the category weight of each device category in the automatic train supervision system comprises: performing weighted summation on the category health degree and the category weight of each device category in the automatic train supervision system to obtain a weighted summation result; determining the health degree of the automatic train supervision system according to the weighted summation result.

8. A health assessment device for an automatic train monitoring system, characterized in that, The method comprises: a first obtaining module configured to obtain multi-dimensional data of each device in an automatic train supervision system, wherein the multi-dimensional data is used to represent running features, historical performance and environment state of the corresponding device; The analysis module is configured to perform health degree analysis on each of the devices based on multi-dimensional data of each of the devices, and obtain initial health degrees of the devices; The second obtaining module is configured to obtain redundancy levels, device categories, and location weight proportions of the devices; The adjustment module is configured to adjust the initial health degrees of the devices according to the redundancy levels of the devices, and obtain adjusted health degrees of the devices; The first determining module is configured to determine category health degrees of the device categories in the automatic train supervision system according to the adjusted health degrees of the devices, the device categories, and the location weight proportions; The second determining module is configured to obtain category weights of the device categories in the automatic train supervision system, and determine a health degree of the automatic train supervision system according to the category health degrees of the device categories in the automatic train supervision system and the category weights.

9. An electronic device comprising a processor and a memory having a computer program stored therein, characterized in that, The computer program is executed by the processor to implement the steps of the health degree evaluation method of the automatic train supervision system according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the health degree evaluation method of the automatic train supervision system according to any one of claims 1 to 7.

Images