Big data management method and system for high-speed rail case maintenance
By real-time monitoring and analysis of node signals and environmental data of high-speed rail chassis, and by utilizing neural networks and damage factor assessment, precise maintenance of high-speed rail chassis has been achieved. This solves the problem of low precision in maintenance management in existing technologies and improves preventive maintenance capabilities and the level of intelligent management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-22
- Publication Date
- 2026-04-07
AI Technical Summary
Existing maintenance management technologies cannot quantitatively decouple the real-time damage level of high-speed rail chassis under different railway network environments, making it difficult to achieve accurate assessment under complex dynamic loads. This results in low accuracy of preventive maintenance decisions and a tendency to misjudge them as hardware damage, leading to undetected fault phenomena.
By monitoring real-time frequency image data of chassis node signals and environmental aerodynamic load data, and using neural network analysis to analyze the physical connection properties of the chassis, combined with micro-damage residual stress calculation and damage factor assessment, the operational reliability is dynamically monitored, enabling precise maintenance and control.
It has improved preventative maintenance capabilities, reduced the probability of false alarms, optimized maintenance strategies, extended the lifespan of key components, and enhanced the level of intelligent management of high-speed rail assets.
Smart Images
Figure CN121809097A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of big data analytics, specifically to a big data management method and system for high-speed rail chassis maintenance. Background Technology
[0002] As a crucial isolation and protection carrier for the train's power conversion and core control system, the operational stability of the high-speed rail chassis directly impacts the operational safety of high-speed railways. Because high-speed trains frequently traverse enclosed spaces such as tunnels during operation, the intense piston effect between the train's front and the tunnel wall when entering a tunnel at ultra-high speed triggers violent atmospheric pressure shock waves and transient negative pressure alternations. This causes the chassis's sealing interface to endure high-frequency and high-amplitude aerodynamic load pulses within a very short period. This cyclical pressure difference creates a physical micro-breathing effect inside and outside the chassis, making it easy for pollutants from the external environment to breach the sealing barrier and be passively drawn into the chassis. When high-speed trains run in tunnels, metal particles from track wear, catenary arc erosion products, and mineral dust left over from tunnel construction cannot effectively disperse over long periods, forming a high concentration of harmful aerosol loads. These pollutants, once inside the chassis, are deposited unevenly on the surfaces of precision circuit modules and electrical connectors due to gravity deposition and electrostatic adsorption caused by the high-frequency mechanical vibrations of the train. This induces complex electrochemical corrosion, oxide layer evolution, and micro-arc discharge phenomena in a sub-optimal state. However, existing maintenance management technologies are mostly based on fixed maintenance cycles or single electrical level monitoring. This not only fails to quantitatively decouple the real-time damage level of the chassis under different network environments but also often misjudges transient disturbances caused by high-frequency vibrations or sudden environmental changes as hardware failures, resulting in numerous undetected faults. Existing data management models struggle to accurately assess the remaining reliability of chassis under complex dynamic loads, severely limiting the accuracy of big data-driven decision-making for preventative maintenance of high-speed rail systems. Summary of the Invention
[0003] The technical problem to be solved by this invention is the low accuracy of big data decision-making in the preventive maintenance of high-speed rail systems in the prior art. It proposes a big data management method and system for high-speed rail chassis maintenance.
[0004] To achieve the above objectives, the technical solution of the big data management method for high-speed rail chassis maintenance of the present invention includes the following steps: S1: Real-time monitoring and extraction of time-frequency image data of the chassis node signals and environmental aerodynamic load data of the operating conditions of the high-speed rail chassis throughout the entire operation cycle; S2: Perform signal feature recognition preprocessing on the time-frequency image data of the chassis node signal, and obtain the basic data of the chassis contact impedance and physical connection attribute data based on the time-frequency image analysis. Import the physical connection attribute data of the chassis into the micro-damage residual stress calculation strategy to calculate the structural stress evolution difference in real time. S3: Analyze the external airflow aerosol composition and instantaneous pressure pulse frequency data of the operating environment of the chassis, and import the instantaneous pressure pulse frequency data into the chassis microenvironment damage impact calculation strategy to calculate auxiliary damage factors in real time. S4: During the operation cycle of high-speed trains, a fixed operating mileage interval is set as a chassis state evolution monitoring unit time period; the basic data of chassis contact impedance and the airflow characteristic response distribution of the operating environment are extracted, and the operational reliability probability of the chassis system is calculated based on the data; S5: Synchronously predict and evaluate the precise maintenance and control requirements of each chassis status evolution monitoring unit during the specified time period; S6: Dynamically monitor the operational reliability probability and precise maintenance and control requirements of each chassis status evolution monitoring unit within a specified time period, execute corresponding maintenance strategies, and record and upload chassis health mode verification data to the high-speed rail operation and maintenance big data management center.
[0005] Preferably, in S1, the time-frequency image data of the chassis node signal includes: node high-frequency sampling pulse image data and mutual inductance distributed impedance time-frequency data; In S2, the preprocessing of the time-frequency image data of the chassis node signals for signal feature recognition, and the acquisition of chassis physical connection attribute data based on time-frequency image analysis, includes the following specific steps: The acquisition of the chassis physical connection attribute data includes the following steps: S211: Convert the high-frequency sampling pulse image data of the chassis contacts into a complex impedance distribution spectrum model and input it into the input layer of the diagnostic neural network. Establish a topological coordinate system with the physical geometric center of the chassis terminals as the origin. S212: Topological features are extracted from the sampled pulse image data of the chassis through convolutional layers. The convolutional layers are used to extract micro-arc features and magnetotropic softening features in the pulse image data. Each convolutional layer includes 10 dynamic weighted convolutional kernels.
[0006] Preferably, in step S2, obtaining the chassis physical connection attribute data further includes the following specific steps: S213: The micro-arc deposition region, material softening void region and electromagnetic penetration region in the port pulse image data output by the convolutional layer are sampled and extracted by the sampling layer, and the features of each region are labeled as different maintenance alarm modes according to the logical feature space of the micro-arc region, material void and coupling noise region. S214: Simultaneously, the correlation attenuation information of the micro-arc region, the material softening void region, and the coupled electromagnetic noise region is compressed into a one-dimensional structural sensitive feature vector, and finally output through the output layer to complete the training of the system health evolution model. S215: Extract the contact surface degradation mapping data, electromagnetic shielding effectiveness evolution data, impedance nonlinear feedback data, and connector mechanical preload stress data output from the neural network model; the contact surface degradation mapping data includes: the total number of oxide layer coverages, the total area of micro-pitting corrosion regions, and the average contact resistance of the signal path.
[0007] Preferably, in step S2, the physical connection attribute data of the chassis is imported into the structural stress evolution calculation strategy, and the chassis structural stress evolution difference is calculated. The calculation strategy is as follows: ; in, These are the health percentage coefficients for the chassis contact surface layer and the structural integrity layer, respectively. This is the assessment value for contact surface degradation; This refers to the health assessment value for the integrity of the chassis structure. The contact surface degradation assessment value The calculation strategy is as follows: ; in, This is the increment for the number of surface pitting defects detected in the image; To measure the area of the oxide layer, This represents the total surface area of the contacts. The oxidation activation energy of the material Here, T is the universal gas constant, and T is the ambient temperature. Weights are balanced by magnitude; The chassis structural integrity health assessment value The calculation strategy is as follows: ; in, This refers to the real-time resonant frequency of the chassis connection structure. This refers to the inherent frequency under design conditions. The standard deviation of grayscale in the contact image. This represents the average gray level.
[0008] Preferably, in step S3, the auxiliary damage factor calculation strategy is as follows: ; in, Environmental load sensitivity coefficient; This represents the real-time air aerosol mass concentration. The effective equivalent area of the chassis sealing gap; and These are the external environmental pressure and the internal respiratory pressure; air density; This refers to the unit monitoring duration.
[0009] Preferably, step S4 includes the following specific steps: S41: In the train's operational lifecycle, a maintenance and management unit period is defined as 5,000 kilometers of cumulative operation. The entire overhaul lifecycle is defined as consisting of DW unit periods. S42: Synchronously extract the fatigue contact impedance of the chassis and the corresponding right-of-way environment data within this cycle; S43: Calculate the operational reliability and health probability of the chassis maintaining high stability during this period based on the data. The operational reliability and health probability The calculation formula is: ; in, These are the hardware quality baseline weight and the environmental degradation correction weight, respectively. This represents the current measured contact impedance of the chassis. This is the factory standard impedance.
[0010] Preferably, in step S5, the calculation formula for the precise maintenance and control requirements is as follows: ; in, These are the control demand weighting coefficients for the structural stress evolution term and its corresponding particulate pollution damage term caused by negative pressure air intake in the external environment.
[0011] Preferably, S6 includes the following specific steps: S61: Dynamically monitor the operational reliability and health probability of each operation and maintenance unit within a specified time interval. When the probability of reliable and healthy operation is less than or equal to the pre-repair threshold, the system will generate a fault dwell alarm, and the deteriorated oxide layer at the interface of the docking plug will be physically dissolved, the material layer cleaned, or the component will be forcibly replaced. Conversely, if the operational reliability threshold is met but the health descent gradient rate is too high, proceed to step S62. S62: When precise maintenance is required When the signal fluctuation is less than the inherent damping tolerance of the chassis system structure under real-time monitoring, it is determined to be a non-physical hardware damage type of jitter caused by sudden weather changes or instantaneous tunnel air pressure waves. Only the signal fluctuation characteristics are recorded and stored in the background noise library to achieve fault target filtering and not to execute the shutdown maintenance process. When the demand for precise maintenance exceeds the set threshold for dynamic health correction coefficient, the microstructure aging index and environmental disturbance residual wave terms in the management model are extracted for precise strategy decision-making.
[0012] In addition, the big data management system for high-speed rail chassis maintenance of this invention includes the following modules: The system includes a data extraction module, a structural stress evolution module, a damage assessment module, a health and survival prediction module, a fault location calculation module, and a real-time differential execution interactive terminal. The data extraction module is used to monitor and extract in real time the time-frequency image data of the chassis node signals and the environmental aerodynamic load data of the operating conditions of the high-speed rail chassis throughout the entire operation cycle. The structural stress evolution module is used to perform signal feature recognition preprocessing on the time-frequency image data of the chassis node signal, and to obtain the basic data of the chassis contact impedance and physical connection attribute data based on the time-frequency image analysis. The physical connection attribute data of the chassis is then imported into the micro-damage residual stress calculation strategy to calculate the structural stress evolution difference in real time. The damage assessment module is used to analyze the external airflow aerosol composition and instantaneous pressure pulse frequency data of the operating environment of the chassis, and import the instantaneous pressure pulse frequency data into the chassis's microenvironment damage impact calculation strategy to calculate auxiliary damage factors in real time. The health and survival prediction module is used to extract basic contact impedance data of the chassis and airflow characteristic response distribution of the operating environment, and calculate the operational reliability probability of the chassis system based on the data. The fault location and calculation module is used to synchronously predict and evaluate the precise maintenance and control requirements of each chassis status evolution monitoring unit during a specific time period. The real-time execution interactive terminal is used to dynamically monitor the operational reliability probability and precise maintenance and control requirements of each chassis status evolution monitoring unit within a time period, execute corresponding maintenance strategies, and record and upload chassis health mode verification data to the high-speed rail operation and maintenance big data management center.
[0013] Compared with the prior art, the technical effects of the present invention are as follows: This invention addresses the technical problem of ambiguous damage perception under complex dynamic conditions in traditional scheduled maintenance by deeply exploring the nonlinear mapping relationship between tunnel air pressure shock waves, industrial aerosol loads, and the physical degradation mechanism of the chassis. This allows maintenance decisions to precede physical failures, improving the system's preventative maintenance capabilities and operational reliability during long-distance, cross-regional operation. Secondly, this invention effectively decouples signal fluctuations caused by tunnel negative pressure pulses or transient vibrations from actual physical damage, filtering out numerous false alarms caused by sudden environmental load changes. This significantly reduces the probability of undetected faults in high-speed rail maintenance, avoiding unnecessary train downtime for inspections and redundant waste of maintenance resources. Finally, this invention utilizes big data analysis to transform single maintenance experience into a self-evolving regression operator, enabling maintenance strategies to adaptively optimize and adjust based on the network environment characteristics, aerodynamic distribution, and environmental pollution levels of different sections. This provides precise scientific decision support for large-scale cross-regional operation and maintenance of high-speed railways, significantly extending the service life of key components and comprehensively improving the intelligent management level of high-speed rail assets. Attached Figure Description
[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart illustrating the big data management method for high-speed rail chassis maintenance according to the present invention. Figure 2 This is a schematic diagram of the big data management system for high-speed rail chassis maintenance according to the present invention. Detailed Implementation
[0015] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0016] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0017] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.
[0018] Example 1: like Figure 1 As shown in the figure, the big data management method for high-speed rail chassis maintenance according to an embodiment of the present invention includes the following specific steps: S1: Real-time monitoring and extraction of time-frequency image data of the chassis node signals and environmental aerodynamic load data of the operating conditions of the high-speed rail chassis throughout the entire operation cycle; In S1, the time-frequency image data of the chassis node signal includes: node high-frequency sampling pulse image data and mutual inductance distributed impedance time-frequency data; S2: Perform signal feature recognition preprocessing on the time-frequency image data of the chassis node signal, and obtain the basic data of the chassis contact impedance and physical connection attribute data based on the time-frequency image analysis. Import the physical connection attribute data of the chassis into the micro-damage residual stress calculation strategy to calculate the structural stress evolution difference in real time. In S2, the preprocessing of the time-frequency image data of the chassis node signals for signal feature recognition, and the obtaining of the chassis physical connection attribute data based on the time-frequency image analysis, includes the following specific steps: The acquisition of the chassis physical connection attribute data includes the following steps: S211: Convert the high-frequency sampling pulse image data of the chassis contacts into a complex impedance distribution spectrum model and input it into the input layer of the diagnostic neural network. Establish a topological coordinate system with the physical geometric center of the chassis terminals as the origin. S212: Topological features are extracted from the sampled pulse image data of the chassis through convolutional layers. The convolutional layers are used to extract micro-arc features and magnetotropic softening features in the pulse image data. Each convolutional layer includes 10 dynamic weighted convolutional kernels. The reconstruction threshold logic for the micro-arc feature regions and softened regions of pulse image data in the convolutional layer includes: The formula for threshold segmentation of micro-arc phenomena in convolutional layers is as follows: ; The formula for predicting material softening diffusion in convolutional layers is as follows: ; in, The nth coordinate on the edge of the micro-arc feature region of the i-th convolutional layer is The signal distortion gradient value; These are the weights for geometric spatial distribution and grayscale contrast features, respectively. The electrothermal stress flow value is the value of the m-th pixel coordinate point on the edge of the material softening failure region of the i-th convolutional layer. These are the structural morphology deformation weights and the state energy level evolution weights, respectively. This represents the spatial weight of the nth sampling point in the two-dimensional planar coordinate system of the topological model. Used as a reference radius; The standard deviation of the anomalous Gaussian noise grayscale extracted from the i-th convolutional layer; Use the base grayscale; This is the set of grayscale features of the stress relaxation region extracted from the i-th convolutional layer.
[0019] It should be noted that the distribution of the discharge ablation ring generated by the ionization corrosion (micro-arc) at the contact point is such that the wider the ablation ring, the greater the risk of impedance jump at the terminal. It should also be noted that the softening of the terminal contact caused by high-speed vibration manifests as structural creep caused by thermal stress, which correlates the micro-displacement bias with the contact path length, forming a micro-strain index.
[0020] S213: The micro-arc deposition region, material softening void region and electromagnetic penetration region in the port pulse image data output by the convolutional layer are sampled and extracted by the sampling layer, and the features of each region are labeled as different maintenance alarm modes according to the logical feature space of the micro-arc region, material void and coupling noise region. S214: Simultaneously, the correlation attenuation information of the micro-arc region, the material softening void region, and the coupled electromagnetic noise region is compressed into a one-dimensional structural sensitive feature vector, and finally output through the output layer to complete the training of the system health evolution model. S215: Extract the contact surface degradation mapping data, electromagnetic shielding effectiveness evolution data, impedance nonlinear feedback data, and connector mechanical preload stress data output from the neural network model; the contact surface degradation mapping data includes: the total number of oxide layer coverages, the total area of micro-pitting corrosion regions, and the average contact resistance of the signal path.
[0021] In step S2, the physical connection attribute data of the chassis is imported into the structural stress evolution calculation strategy, and the chassis structural stress evolution difference is calculated. The calculation strategy is as follows: ; in, These are the health percentage coefficients for the chassis contact surface layer and the structural integrity layer, respectively. This is the assessment value for contact surface degradation; This refers to the health assessment value for the integrity of the chassis structure. The contact surface degradation assessment value The calculation strategy is as follows: ; in, This is the increment for the number of surface pitting defects detected in the image; To measure the area of the oxide layer, This represents the total surface area of the contacts. The oxidation activation energy of the material Here, T is the universal gas constant, and T is the ambient temperature. Weights are balanced by magnitude; It should be noted that the surface deterioration assessment value of the contact is obtained by evaluating the increase in the number of pits and the ratio of oxidation area. The more pits there are, the larger the oxidation area is, the higher the temperature is, and the more severe the surface deterioration is. The number of pits is discrete, while the ratio of oxidation area is continuous, so the weighting coefficient needs to be adjusted. The chassis structural integrity health assessment value The calculation strategy is as follows: ; in, This refers to the real-time resonant frequency of the chassis connection structure. This refers to the inherent frequency under design conditions. The standard deviation of grayscale in the contact image. Average gray level; It should be noted that, in this embodiment, the degree of electrical signal disturbance induced by mechanical loosening is described by the product of frequency offset rate and visual disturbance rate; it should also be noted that the chassis structural integrity health assessment value is obtained by evaluating the offset of structural resonant frequency and the non-uniformity of image grayscale (reflecting structural loosening or damage). S3: Analyze the external airflow aerosol composition and instantaneous pressure pulse frequency data of the operating environment of the chassis, and import the instantaneous pressure pulse frequency data into the chassis microenvironment damage impact calculation strategy to calculate auxiliary damage factors in real time. In S3, the auxiliary damage factor calculation strategy is as follows: ; in, Environmental load sensitivity coefficient; This represents the real-time air aerosol mass concentration. The effective equivalent area of the chassis sealing gap; and These are the external environmental pressure and the internal respiratory pressure; air density; For unit monitoring duration; For example, an environmental load sensitivity coefficient is given. The acquisition strategy is as follows: ; The effective deposition rate of aerosols represents the proportion of contaminants (such as salt spray / dust) drawn into the chassis with the aerodynamic load that are deposited and retained on the surface of the internal connectors. This is the environmental corrosion activity gain factor, representing the input based on road network big data. The coefficient for weighting the electrochemical erosion rate by the type of aerosol (such as high salinity aerosol in coastal areas and silicate dust in Gobi areas); The dust-binding failure threshold within a unit life cycle of the chassis; In this embodiment, computational fluid dynamics (CFD) software is used to simulate the internal flow field of the train chassis when it passes through a tunnel at high speed. Based on the grille design of the chassis air inlet and the internal PCB board layout, a three-phase flow model of gas, liquid, and solid is established. The ratio of the mass of particles retained on the connector surface to the total mass of particles in the intake airflow is used as... ; It should be noted that this formula calculates the total cumulative amount of damaging contaminants entering the chassis during the cycle by integrating the intake mass flow induced by aerodynamic load over time, and finally outputs the damage index.
[0022] S4: During the operation cycle of high-speed trains, a fixed operating mileage interval is set as a chassis state evolution monitoring unit time period; the basic data of chassis contact impedance and the airflow characteristic response distribution of the operating environment are extracted, and the operational reliability probability of the chassis system is calculated based on the data; S4 includes the following specific steps: S41: In the train's operational lifecycle, a maintenance and management unit period is defined as 5,000 kilometers of cumulative operation. The entire overhaul lifecycle is defined as consisting of DW unit periods. S42: Synchronously extract the fatigue contact impedance of the chassis and the corresponding right-of-way environment data within this cycle; S43: Calculate the operational reliability and health probability of the chassis maintaining high stability during this period based on the data. The operational reliability and health probability The calculation formula is: ; in, These are the hardware quality baseline weight and the environmental degradation correction weight, respectively. This represents the current measured contact impedance of the chassis. The impedance is the factory standard impedance. S5: Synchronously predict and evaluate the precise maintenance and control requirements of each chassis status evolution monitoring unit during the specified time period; In S5, the calculation formula for the precise maintenance and control requirements is as follows: ; in, These are the control demand weighting coefficients for the structural stress evolution term and its corresponding particulate pollution damage term caused by negative pressure air intake in the external environment.
[0023] S6: Dynamically monitor the operational reliability probability and precise maintenance and control requirements of each chassis status evolution monitoring unit within a specified time period, execute corresponding maintenance strategies, and record and upload chassis health mode verification data to the high-speed rail operation and maintenance big data management center.
[0024] S6 includes the following specific steps: S61: Dynamically monitor the operational reliability and health probability of each operation and maintenance unit within a specified time interval. When the probability of reliable and healthy operation is less than or equal to the pre-repair threshold, the system will generate a fault dwell alarm, and the deteriorated oxide layer at the interface of the docking plug will be physically dissolved, the material layer cleaned, or the component will be forcibly replaced. Conversely, if the operational reliability threshold is met but the health descent gradient rate is too high, proceed to step S62. S62: When precise maintenance is required When the signal fluctuation is less than the inherent damping tolerance of the chassis system structure under real-time monitoring, it is determined to be a non-physical hardware damage type of jitter caused by sudden weather changes or instantaneous tunnel air pressure waves. Only the signal fluctuation characteristics are recorded and stored in the background noise library to achieve fault target filtering and not to execute the shutdown maintenance process. When the demand for precise maintenance exceeds the set threshold for dynamic health correction coefficient, the microstructure aging index and environmental disturbance residual effect items in the management model are extracted to make targeted actuarial strategy decisions.
[0025] S63: Select the feature module with the largest difference in physical stress evolution for maintenance decision: When the cumulative slope of the nonlinear impedance offset generated by the oxide wear layer reaches the preset chemical activation energy evolution threshold, send a lubrication balance treatment to the high-frequency oscillating component. S64: Select environmental risk indicators based on road network damage factors for anti-penetration treatment: When the auxiliary damage factor caused by the cumulative inhalation mass of external aerosols reaches the safety limit threshold, execute the anti-aerodynamic stress optimization plan based on big data historical records to reinforce the shell sealing strips of vehicles operating in this area with severe pressure fluctuations.
[0026] S65: After all physical safety indicators are restored to the equilibrium range of the rated physical safety threshold, the full-frequency impedance characteristic spectrum comparison data of the contacts before and after this maintenance is summarized to form a maintenance management report, which is then transmitted to the headquarters machine maintenance center as the regression operator for the next high-speed rail element operation health weight prediction.
[0027] Example 2: like Figure 2 As shown in the figure, the big data management system for high-speed rail chassis maintenance according to an embodiment of the present invention is as follows: Figure 2 As shown, it includes the following modules: The system includes a data extraction module, a structural stress evolution module, a damage assessment module, a health and survival prediction module, a fault location calculation module, and a real-time differential execution interactive terminal. The data extraction module is used to monitor and extract in real time the time-frequency image data of the chassis node signals and the environmental aerodynamic load data of the operating conditions of the high-speed rail chassis throughout the entire operation cycle. The structural stress evolution module is used to perform signal feature recognition preprocessing on the time-frequency image data of the chassis node signal, and to obtain the basic data of the chassis contact impedance and physical connection attribute data based on the time-frequency image analysis. The physical connection attribute data of the chassis is then imported into the micro-damage residual stress calculation strategy to calculate the structural stress evolution difference in real time. The damage assessment module is used to analyze the external airflow aerosol composition and instantaneous pressure pulse frequency data of the operating environment of the chassis, and import the instantaneous pressure pulse frequency data into the chassis's microenvironment damage impact calculation strategy to calculate auxiliary damage factors in real time. The health and survival prediction module is used to extract basic contact impedance data of the chassis and airflow characteristic response distribution of the operating environment, and calculate the operational reliability probability of the chassis system based on the data. The fault location and calculation module is used to synchronously predict and evaluate the precise maintenance and control requirements of each chassis status evolution monitoring unit during a specific time period. The real-time execution interactive terminal is used to dynamically monitor the operational reliability probability and precise maintenance and control requirements of each chassis status evolution monitoring unit within a time period, execute corresponding maintenance strategies, and record and upload chassis health mode verification data to the high-speed rail operation and maintenance big data management center.
[0028] Example 3: This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the aforementioned big data management method for high-speed rail chassis maintenance by calling computer programs stored in memory.
[0029] The electronic device can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the big data management method for high-speed rail chassis maintenance provided in the above-described embodiment. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Further details are omitted in this embodiment.
[0030] Example 4: This embodiment proposes a computer-readable storage medium on which an erasable and rewritable computer program is stored. When the computer program runs on the computer device, it causes the computer device to execute the big data management method described above for high-speed rail chassis maintenance.
[0031] For example, computer-readable storage media can be read-only memory (ROM), random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage devices.
[0032] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0033] It should be understood that determining B based on A does not mean determining B solely based on A; it also means determining B based on A and / or other information.
[0034] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired network and / or wireless network. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives (SSDs).
[0035] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0036] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.
Claims
1. A big data management method for high-speed rail chassis maintenance, characterized in that, The method includes: S1: Real-time monitoring and extraction of time-frequency image data of the chassis node signals and environmental aerodynamic load data of the operating conditions of the high-speed rail chassis throughout the entire operation cycle; S2: Perform signal feature recognition preprocessing on the time-frequency image data of the chassis node signal, and obtain the basic data of the chassis contact impedance and physical connection attribute data based on the time-frequency image analysis. Import the physical connection attribute data of the chassis into the micro-damage residual stress calculation strategy to calculate the structural stress evolution difference in real time. S3: Analyze the external airflow aerosol composition and instantaneous pressure pulse frequency data of the operating environment of the chassis, and import the instantaneous pressure pulse frequency data into the chassis microenvironment damage impact calculation strategy to calculate auxiliary damage factors in real time. S4: During the operation cycle of high-speed trains, a fixed operating mileage interval is set as a chassis state evolution monitoring unit time period; the basic data of chassis contact impedance and the airflow characteristic response distribution of the operating environment are extracted, and the operational reliability probability of the chassis system is calculated based on the data; S5: Synchronously predict and evaluate the precise maintenance and control requirements of each chassis status evolution monitoring unit during the specified time period; S6: Dynamically monitor the operational reliability probability and precise maintenance and control requirements of each chassis status evolution monitoring unit within a specified time period, execute corresponding maintenance strategies, and record and upload chassis health mode verification data to the high-speed rail operation and maintenance big data management center.
2. The big data management method for high-speed rail chassis maintenance according to claim 1, characterized in that, In S1, the time-frequency image data of the chassis node signal includes: node high-frequency sampling pulse image data and mutual inductance distributed impedance time-frequency data; In S2, the preprocessing of the time-frequency image data of the chassis node signals for signal feature recognition, and the acquisition of chassis physical connection attribute data based on time-frequency image analysis, includes the following specific steps: The acquisition of the chassis physical connection attribute data includes the following steps: S211: Convert the high-frequency sampling pulse image data of the chassis contacts into a complex impedance distribution spectrum model and input it into the input layer of the diagnostic neural network. Establish a topological coordinate system with the physical geometric center of the chassis terminals as the origin. S212: Topological features are extracted from the sampled pulse image data of the chassis through convolutional layers. The convolutional layers are used to extract micro-arc features and magnetotropic softening features in the pulse image data. Each convolutional layer includes 10 dynamic weighted convolutional kernels.
3. The big data management method for high-speed rail chassis maintenance according to claim 2, characterized in that, In S2, obtaining the chassis physical connection attribute data further includes the following specific steps: S213: The micro-arc deposition region, material softening void region and electromagnetic penetration region in the port pulse image data output by the convolutional layer are sampled and extracted by the sampling layer, and the features of each region are labeled as different maintenance alarm modes according to the logical feature space of the micro-arc region, material void and coupling noise region. S214: Simultaneously, the correlation attenuation information of the micro-arc region, the material softening void region, and the coupled electromagnetic noise region is compressed into a one-dimensional structural sensitive feature vector, and finally output through the output layer to complete the training of the system health evolution model. S215: Extract the contact surface degradation mapping data, electromagnetic shielding effectiveness evolution data, impedance nonlinear feedback data, and connector mechanical preload stress data output from the neural network model. The contact surface degradation mapping data includes: the total number of oxide layer coverages, the total area of micro-pitting regions, and the average contact resistance of the signal path.
4. The big data management method for high-speed rail chassis maintenance according to claim 3, characterized in that, In step S2, the physical connection attribute data of the chassis is imported into the structural stress evolution calculation strategy, and the chassis structural stress evolution difference is calculated. The calculation strategy is as follows: ； in, These are the health percentage coefficients for the chassis contact surface layer and the structural integrity layer, respectively. This is the assessment value for contact surface degradation; This refers to the health assessment value for the integrity of the chassis structure. The contact surface degradation assessment value The calculation strategy is as follows: ； in, This is the increment for the number of surface pitting defects detected in the image; To measure the area of the oxide layer, This represents the total surface area of the contacts. The oxidation activation energy of the material Here, T is the universal gas constant, and T is the ambient temperature. Weights are balanced by magnitude; The chassis structural integrity health assessment value The calculation strategy is as follows: ； in, This refers to the real-time resonant frequency of the chassis connection structure. This refers to the inherent frequency under design conditions. The standard deviation of grayscale in the contact image. This represents the average gray level.
5. The big data management method for high-speed rail chassis maintenance according to claim 4, characterized in that, In S3, the auxiliary damage factor calculation strategy is as follows: ； in, Environmental load sensitivity coefficient; This represents the real-time air aerosol mass concentration. The effective equivalent area of the chassis sealing gap; and These are the external environmental pressure and the internal respiratory pressure; air density; This refers to the unit monitoring duration.
6. The big data management method for high-speed rail chassis maintenance according to claim 5, characterized in that, S4 includes the following specific steps: S41: In the train's operational lifecycle, a maintenance and management unit period is defined as 5,000 kilometers of cumulative operation. The entire overhaul lifecycle is defined as consisting of DW unit periods. S42: Synchronously extract the fatigue contact impedance of the chassis and the corresponding right-of-way environment data within this cycle; S43: Calculate the operational reliability and health probability of the chassis maintaining high stability during this period based on the data. The operational reliability and health probability The calculation formula is: ； in, These are the hardware quality baseline weight and the environmental degradation correction weight, respectively. This represents the current measured contact impedance of the chassis. This is the factory standard impedance.
7. The big data management method for high-speed rail chassis maintenance according to claim 6, characterized in that, In S5, the calculation formula for the precise maintenance and control requirements is as follows: ； in, These are the control demand weighting coefficients for the structural stress evolution term and its corresponding particulate pollution damage term caused by negative pressure air intake in the external environment.
8. The big data management method for high-speed rail chassis maintenance according to claim 7, characterized in that, S6 includes the following specific steps: S61: Dynamically monitor the operational reliability and health probability of each operation and maintenance unit within a specified time interval. ; When the probability of reliable and healthy operation is less than or equal to the pre-repair threshold, the system is triggered to generate a fault dwell alarm, and the deteriorated oxide layer at the interface of the docking plug is physically dissolved, the material layer is cleaned, or the component is forcibly replaced. Conversely, if the operational reliability threshold is met but the health descent gradient rate is too high, proceed to step S62. S62: When precise maintenance is required When the signal fluctuation is less than the inherent damping tolerance of the chassis system structure under real-time monitoring, it is determined to be a non-physical hardware damage type of jitter caused by sudden weather changes or instantaneous tunnel air pressure waves. Only the signal fluctuation characteristics are recorded and stored in the background noise library to achieve fault target filtering and not to execute the shutdown maintenance process. When the demand for precise maintenance exceeds the set threshold for dynamic health correction coefficient, the microstructure aging index and environmental disturbance residual wave terms in the management model are extracted for precise strategy decision-making.
9. A big data management system for high-speed rail chassis maintenance, used to implement the big data management method for high-speed rail chassis maintenance as described in any one of claims 1-8, characterized in that, The system includes the following modules: The system includes a data extraction module, a structural stress evolution module, a damage assessment module, a health and survival prediction module, a fault location calculation module, and a real-time differential execution interactive terminal. The data extraction module is used to monitor and extract in real time the time-frequency image data of the chassis node signals and the environmental aerodynamic load data of the operating conditions of the high-speed rail chassis throughout the entire operation cycle. The structural stress evolution module is used to perform signal feature recognition preprocessing on the time-frequency image data of the chassis node signal, and to obtain the basic data of the chassis contact impedance and physical connection attribute data based on the time-frequency image analysis. The physical connection attribute data of the chassis is then imported into the micro-damage residual stress calculation strategy to calculate the structural stress evolution difference in real time. The damage assessment module is used to analyze the external airflow aerosol composition and instantaneous pressure pulse frequency data of the operating environment of the chassis, and import the instantaneous pressure pulse frequency data into the chassis's microenvironment damage impact calculation strategy to calculate auxiliary damage factors in real time. The health and survival prediction module is used to extract basic contact impedance data of the chassis and airflow characteristic response distribution of the operating environment, and calculate the operational reliability probability of the chassis system based on the data. The fault location and calculation module is used to synchronously predict and evaluate the precise maintenance and control requirements of each chassis status evolution monitoring unit during a specific time period. The real-time execution interactive terminal is used to dynamically monitor the operational reliability probability and precise maintenance and control requirements of each chassis status evolution monitoring unit within a time period, execute corresponding maintenance strategies, and record and upload chassis health mode verification data to the high-speed rail operation and maintenance big data management center.