Subway vehicle abnormal jitter repairing method, system, equipment and medium

Abstract

The invention belongs to the technical field of urban rail transit operation and maintenance, and discloses a metro vehicle abnormal jitter repairing method, system, device and medium, and the method comprises the steps: 1, collecting a vehicle body vibration signal, fusing passenger comfort feedback, and constructing a vehicle body abnormal jitter grading system related to objective indexes and subjective feelings; secondly, collecting static environment parameters and vehicle running dynamic environment parameters around the track, building a multivariable regression model, and quantifying the influence weight of each parameter on jitter; thirdly, main environmental factors are found out according to the influence weights, key vibration transmission points are determined through simulation of a vehicle body-track rigid-flexible coupling model and energy flow analysis of a vehicle body-bogie finite element model, and a fault source is positioned; and 4, generating a temporary protection and warehousing repair scheme according to the fault source, the temporary protection scheme being a speed adjustment curve optimized by a genetic algorithm, and the warehousing repair scheme being a personalized operation instruction based on a mapping library. The method can accurately diagnose the abnormal jitter reason of the vehicle body and generate an efficient repair scheme.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of urban rail transit operation and maintenance, and particularly relates to a subway vehicle abnormal shaking repair method, system, device and medium. BACKGROUND

[0002] As the core carrier of urban public transportation, the running stability and riding comfort of the subway directly relate to the travel experience and life safety of passengers. With the continuous increase of operation mileage and the extension of service time, under the influence of multiple factors such as complex wheel-rail interaction, track environment disturbance and component aging, the abnormal shaking of the subway vehicle body is increasingly prominent. This non-periodic and sudden vibration not only causes passengers to feel uncomfortable, leading to passenger complaints and increasing operating costs, but also accelerates the fatigue damage of key components such as axle box bearings and suspension systems, leading to increased maintenance costs, and even induces deterioration of wheel-rail relationship, increases the risk of derailment and other safety hazards. Therefore, the accurate diagnosis and efficient repair of the abnormal shaking of the urban rail vehicle body have become an urgent need in the field of rail transit operation and maintenance.

[0003] However, the current monitoring and processing technology for abnormal shaking of the subway vehicle body still has significant deficiencies: 1. The existing technology mainly relies on a single physical parameter, such as the peak value of the vertical acceleration of the vehicle body, to judge the shaking state, without considering the correlation between the time-frequency characteristics of the vibration signal and environmental factors and the subjective feelings of passengers. In addition, the low-frequency components contained in the vibration signal mainly affect the riding comfort, while the high-frequency components are more likely to reflect component failures. The existing technology does not subdivide the frequency bands of the signal, resulting in one-sided diagnosis results. Relying solely on acceleration peak values may misjudge low-frequency shaking caused by track settlement as wheelset imbalance, causing waste of maintenance resources. Moreover, shaking, as a phenomenon that directly affects the riding experience, its severity not only depends on physical parameters, but also closely related to the subjective feelings of passengers. The existing technology relies entirely on sensor data without incorporating passenger feedback information. For example, some passengers are more sensitive to low-frequency vibrations between 1-3Hz, but the existing grading system does not reflect this difference, resulting in some shaking events that significantly affect specific groups being missed. Many passenger complaints involve shaking events that do not meet the traditional warning thresholds, reflecting the disconnection between objective indicators and subjective feelings.

[0004] 2. Temporary protective measures lack dynamic optimization. When severe shaking is detected, the existing technology usually adopts a fixed deceleration scheme without optimizing it in combination with line scheduling, passenger flow peaks and other actual situations. This "one-size-fits-all" approach often leads to unnecessary operational delays, such as implementing significant speed reduction on key lines during the morning rush hour, which may cause passenger flow congestion throughout the network. By optimizing the speed curve, the delay time can be shortened, reflecting the optimization space for temporary measures.

[0005] 3. Repair the missing closed-loop mechanism, and the fault is easy to repeat. After completing the jitter repair, the existing technology lacks systematic effect verification and long-term tracking. Some maintenance only targets surface phenomena and does not address the root cause, leading to the recurrence of the same jitter problem in the short term. In addition, the jitter data between different lines is not shared, and a cross-line fault knowledge base cannot be formed, leading to the recurrence of common problems that cannot be systematically solved.

[0006] Therefore, there is a lack of a systematic, data-driven closed-loop method that can run through the whole process of "perception-diagnosis-decision-repair-verification" in existing methods to solve the problem of inaccurate fault positioning, low repair efficiency and repeated problems caused by data chain breakage of abnormal jitter of the vehicle body. SUMMARY

[0007] The purpose of the present application is to solve the core problem of inaccurate fault positioning, low repair efficiency and repeated problems caused by the disconnection of the data chain in the whole process of diagnosis, decision and repair verification in existing methods. To this end, the present application provides a subway vehicle abnormal jitter repair method, system, device and medium.

[0008] In a first aspect, the present application provides a subway vehicle abnormal jitter repair method, comprising the following steps: S1, collecting the vehicle body vibration signal and fusing the passenger comfort feedback information to establish a vehicle abnormal jitter grading system that associates objective vibration indicators with subjective passenger experience; S2, collecting static environmental parameters of the track surrounding environment and dynamic environmental parameters of the vehicle operation, establishing a multivariate regression model of environmental parameters and vehicle jitter, and quantifying the influence weight of each environmental parameter on jitter; S3, based on the main environmental impact factors identified by the influence weight, simulating the vibration transmission characteristics through the vehicle-body-track rigid-flexible coupling model, and combining the vehicle-body-bogie finite element model for energy flow analysis to determine the key vibration transmission point causing abnormal jitter to locate the fault source; S4, generating temporary protective measures and in-storage repair schemes according to the located fault source, the temporary protective measures being a speed adjustment curve generated by combining a genetic algorithm optimization, and the in-storage repair scheme being an individualized job instruction sheet generated based on a fault-repair process mapping library.

[0009] As an optional implementation form of the first aspect of the application, step S1 specifically comprises: arranging ICP type three-axis acceleration sensors on the top of the front end of the driver's cab of the vehicle body, the floor on both sides of the middle part of the vehicle body, the side walls at the front and rear ends of the passenger compartment, and the bottom of the vehicle body above the bogie; performing high-pass filtering on the collected original vibration signals to remove the direct current component, performing Butterworth low-pass filtering to suppress high-frequency noise, and eliminating impulse noise using the 3σ criterion; calculating the acceleration root mean square value RMS and the peak factor CF of the preprocessed signals, and identifying the main vibration frequency components, and constructing a comprehensive vibration index CVI as an objective vibration index through weighted summation.

[0010] As an optional implementation form of the first aspect of the application, step S1 further specifically comprises: collecting the 1-5 point rating of passengers on the vibration intensity, vibration frequency, duration, standing stability, and overall comfort of the five core indicators through a vehicle-mounted application program or a two-dimensional code; performing weighted summation on the five indicator ratings of a single feedback to construct an individual comfort index ICI; performing statistics on the individual comfort indexes ICI of all feedbacks of the same train to construct a group comfort index GCI; and establishing a shaking classification system including five levels of slight, mild, moderate, severe, and extremely severe based on the comprehensive vibration index CVI and the group comfort index GCI as two dimensions.

[0011] As an optional implementation form of the first aspect of the application, steps S2 and S3 specifically comprise: the collected static environment parameters include the track foundation settlement measured by a total station, the tunnel structure deformation collected by a shield segment joint displacement meter, the rail profile measured by a laser profiler, and the track bed stiffness measured by a falling weight deflectometer; the collected dynamic environment parameters include the train running speed calculated by a vehicle-mounted GPS and wheel diameter, the passenger capacity counted based on a vehicle door infrared sensor, track curve parameters, and temperature in the tunnel; the Simpack software is used to establish the rigid-flexible coupling model of the vehicle body and the track, wherein the vehicle subsystem includes a flexible vehicle body, a rigid bogie, a flexible wheelset, and a suspension system, the track subsystem includes flexible rails, sleepers, and a track bed, the wheel-rail contact uses a Hertz nonlinear contact to calculate the normal force, and the KalkerFastsim algorithm is used to calculate the tangential force; the Abaqus software is used to construct the finite element model of the vehicle body and the bogie, the measured wheel-rail excitation force is applied for transient dynamics analysis, and the energy flow analysis method is used to calculate the vibration energy transmission rate of each connection node, and the node with the highest vibration energy transmission rate is taken as the key vibration transmission point.

[0012] As an optional implementation form of the first aspect of the present application, in step S4, the generation of the speed adjustment curve comprises: constructing a multi-objective optimization function with the target of running time extension, vibration exceeding times and energy consumption increment; setting constraint conditions, which include: safety constraints, wherein the maximum deceleration acceleration is not greater than 0.5 m / s², the minimum running speed of the straight line segment is not less than 20 km / h, and the minimum running speed of the curve segment is not less than 10 km / h; time constraints, wherein the single trip delay time is not greater than 5 minutes; and line constraints, wherein the speed limit values under different curve and slope conditions are included; solving the multi-objective optimization function under the constraint conditions by using a genetic algorithm to obtain the optimal speed adjustment curve, and pushing the speed adjustment curve to the dispatch center through a train-ground communication system.

[0013] As an optional implementation form of the first aspect of the present application, step S4 specifically comprises: establishing the fault-repair process mapping library which divides the faults into 6 categories and 32 subcategories, including wheel-rail contact category, suspension system category, track structure category, vehicle body structure category, environmental influence category and other category; formulating a standard repair process including required tool list, key process parameters and quality inspection standards for each category of fault in the library; generating the individualized job instruction book based on the BIM model, which contains a three-dimensional schematic diagram of the fault position, step-by-step operation instructions, matters needing attention, and work time and resource estimation.

[0014] As an optional implementation form of the first aspect of the present application, the method further comprises the following steps: S5, after executing the warehouse repair scheme, performing a trial run, collecting vibration data after repair and comparing with the vibration data before repair, and quantitatively evaluating the repair effect; wherein the indicators for quantitatively evaluating the repair effect include: vibration improvement rate, wherein the vibration improvement rate is not less than 70% for 3-level and above vibration, and not less than 50% for 1-2 level vibration; comfort degree improvement degree, wherein the difference of group comfort index GCI before and after repair is not less than 1.0; and key component state, wherein the stiffness deviation of primary spring is not greater than 3%, and the wheel set diameter difference is not greater than 0.3 mm; S6, establishing a long-term tracking mechanism for the repaired vehicle, and constructing a cross-line big data platform for data sharing and common problem analysis; wherein the construction of the cross-line big data platform comprises: collecting vibration data once a week to calculate the weekly average of the comprehensive vibration index CVI, and issuing a warning when the CVI rising amplitude is not less than 10% for three consecutive weeks; predicting the remaining life of the component based on the vibration history of the key component by using a fatigue cumulative damage model; identifying common problems and revising the vehicle maintenance standard by using K-means algorithm for cluster analysis of fault data of multiple lines; and constructing a forward-looking early warning model by using LSTM neural network, inputting long-term tracking data, environmental parameters and operation data, predicting the vibration level in the next three months, and formulating preventive maintenance measures in advance when the predicted vibration level reaches 3 level and above.

[0015] In a second aspect, the embodiments of the present application provide a subway vehicle abnormal shaking repair system, comprising: a vehicle abnormal shaking grading module configured to collect vehicle vibration signals and fuse passenger comfort feedback information, and establish a vehicle abnormal shaking grading system associating objective vibration indicators with subjective passenger experience; a track environment parameter quantification module configured to collect static environment parameters of a track surrounding environment and dynamic environment parameters of vehicle operation, establish a multivariate regression model of environment parameters and vehicle shaking, and quantify the influence weight of each environment parameter on shaking; an environment-shaking correlation analysis fault source module configured to identify main environment influencing factors based on the influence weight, simulate vibration transmission characteristics through a vehicle-track rigid-flexible coupling model, and combine a vehicle-bogie finite element model to perform energy flow analysis, determine a key vibration transmission point causing abnormal shaking to locate a fault source; an intelligent decision-making and repair strategy generation module configured to generate temporary protective measures and in-storage repair schemes according to the located fault source, wherein the temporary protective measures are speed adjustment curves generated by combining a genetic algorithm optimization, and the in-storage repair schemes are individualized job instruction sheets generated based on a fault-repair process mapping library.

[0016] In a third aspect, the embodiments of the present application provide an electronic device, which comprises a processor, a memory, and a program or instruction stored on the memory and executable on the processor, and the program or instruction is executed by the processor to implement the steps of the method according to the first aspect.

[0017] In a fourth aspect, the embodiments of the present application provide a readable storage medium, which stores a program or instruction, and the program or instruction is executed by a processor to implement the steps of the method according to the first aspect.

[0018] Compared with the prior art, the present application realizes accurate diagnosis and efficient repair of abnormal shaking of the subway vehicle body, which is due to the fact that the present application establishes a hierarchical system by fusing passenger comfort feedback (step S1) through multi-source data collection and quantitative index construction (step S2), thereby comprehensively depicting the shaking state and breaking away from the limitations of traditional single physical parameter judgment, and improving the accuracy of diagnosis. At the same time, the present application establishes a multi-dimensional fault tracing mechanism by coupling static and dynamic environmental parameters through environment-shaking correlation analysis (step S3), and locating the fault source by combining the vehicle-track rigid-flexible coupling model and finite element analysis, thereby significantly improving the accuracy of fault location. In addition, by optimizing the temporary protective measures through genetic algorithm, the present application generates an individualized in-storage repair scheme (step S4), and relying on the closed-loop feedback mechanism and long-term tracking and cross-line big data platform (step S5), the present application effectively balances the operation safety and efficiency, handles the dynamics and complexity in the repair process, and greatly improves the effectiveness and long-term effectiveness of the repair. Finally, through the systematic diagnosis, repair and management process, the present application can accurately guide the vehicle maintenance decision, optimize the traditional maintenance mode, avoid resource waste, significantly reduce the maintenance cost, and fundamentally improve the safety and comfort of subway operation, realizing the change from passive response to active prevention and precise maintenance. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a technical roadmap of a subway vehicle abnormal shaking repair method according to an embodiment of the present application; Figure 2 is a vehicle-track rigid-flexible coupling dynamics model schematic diagram according to an embodiment of the present application; Figure 3 is a subway vehicle bogie and vehicle body connection topology diagram according to an embodiment of the present application; Figure 4 is a fault type and repair process mapping logic diagram according to an embodiment of the present application; Figure 5 is a structural schematic diagram of a subway vehicle abnormal shaking repair system provided by an embodiment of the present application.

[0020] The following specific embodiments will further illustrate the present application in conjunction with the above drawings. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0022] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0023] Example 1 Please see Figure 1 This is a flowchart illustrating a method for repairing abnormal vibrations in subway vehicles, provided by an embodiment of the present invention. The method includes the following steps: S1. Collect vehicle vibration signals and integrate passenger comfort feedback information to establish a vehicle abnormal vibration classification system that correlates objective vibration indicators with subjective passenger feelings.

[0024] I. Sensor Selection and Arrangement Sensor type: ICP type triaxial accelerometer is selected to meet the wideband measurement requirements of subway vibration signals; Layout plan as follows Figure 2 Adopting the principle of "full coverage of key nodes", eight sensors are arranged on the car body structure: the top of the front of the driver's cab (monitoring longitudinal vibration); the floor on both sides of the middle of the car body (monitoring vertical and lateral vibration); the front and rear side walls of the passenger compartment (monitoring vibration in the passenger area); and the bottom of the car body above the bogie (monitoring vibration transmitted by the suspension system). Installation requirements: The sensor is rigidly fixed to the vehicle body steel structure with bolts, and the flatness of the mounting surface is ≤0.05mm to ensure that the vibration signal is transmitted without attenuation.

[0025] II. Signal Preprocessing DC component removal: A high-pass filter (cutoff frequency 0.5Hz) is used to eliminate the effects of sensor zero drift and temperature drift. The formula is as follows: ,in, The original signal, The signal after removing the DC component. T The sampling period.

[0026] Noise suppression: A 5th-order Butterworth low-pass filter (cutoff frequency 100Hz) is used to filter out high-frequency noise. The filter function is: ,in, The cutoff angular frequency is (2π×100rad / s), and n=5 is the filter order.

[0027] Impulse noise cancellation: outlier identification using the 3σ criterion ( (Identified as impulse noise), replaced by linear interpolation, the formula is: ,in, The mean of the signal. Standard deviation This is the corrected value.

[0028] III. Calculation of Quantitative Indicators Root mean square value of acceleration ( RMS ): Reflects the overall level of vibrational energy, the formula is: ,in, This is the preprocessed acceleration value. N This represents the number of sampling points.

[0029] Peak factor (CF): reflects the intensity of impact vibration, and the formula is: Identification of primary vibration frequencies: Divide the 0-100Hz frequency band into 20 1 / 3 octave bands, calculate the energy percentage of each band, and identify the primary vibration frequency components. The formula is as follows: ,in For the first k The frequency range of each frequency band This is the result of the Fourier transform of the signal.

[0030] Indicator Fusion: A comprehensive vibration index is constructed using weighted summation. CVI The formula is: The weights are determined using the analytic hierarchy process (AHP) to reflect the contribution of each indicator to the perception of jitter.

[0031] IV. Passenger Feedback Data Collection Feedback channels: Two feedback methods have been developed: an in-vehicle app and a QR code scanning system, ensuring convenient participation for passengers of all ages; Feedback content: Includes 5 core indicators (rating from 1 to 5 points): Vibration intensity (1 = No sensation, 2 = Slight sensation, 3 = Sensation, 4 = Slightly strong sensation, 5 = Unbearable). Vibration frequency (1 = Low-frequency stable, 2 = Slightly unstable, 3 = Unstable, 4 = Very unstable, 5 = Unbearable). Duration (1 = instantaneous, 2 = at least 3 seconds, 3 = at least 1 minute, 4 = at least a few minutes, 5 = continuous throughout). Standing stability (1 = No effect, 2 = Feel the sway, 3 = Some sway, 4 = Noticeable sway, 5 = Difficult to stand); Overall comfort (1 = Comfortable, 2 = Somewhat uncomfortable, 3 = Obviously uncomfortable, but intend to stay until the destination, 4 = Very uncomfortable, intend to get off at the next stop, 5 = Extremely uncomfortable, wish for the train to stop now). Sampling strategy: Collect ≥50 feedback samples per train per operating day. If insufficient, supplement with a weighted algorithm (weights are positively correlated with travel duration).

[0032] V. Quantitative Processing of Feedback Data Individual comfort index ( ICI The formula for the weighted sum of the five indicators in a single feedback is: ,in, These are the scores for each indicator.

[0033] Group comfort index ( GCI ): All feedback for the same train number ICI The statistical formula is as follows: ,in, For the first i Weighting of feedback (when travel time is >30 minutes) ,otherwise ), M This refers to the number of feedback copies.

[0034] VI. Construction of Dynamic Hierarchical System Grading index: based on the comprehensive vibration index ( CVI ) and group comfort index ( GCI A two-dimensional, five-level grading standard is established: Level 1 (Slight): and The passengers experienced no obvious discomfort. Level 2 (Mild): and A few sensitive passengers felt it; Level 3 (Moderate): and Most passengers felt it clearly; Level 4 (Severe): and Passengers were generally uncomfortable; Level 5 (Extremely Severe): and This seriously affects passenger safety; Dynamic adjustment mechanism: The classification thresholds are updated quarterly based on the distribution characteristics of passenger feedback to ensure consistency between the classification results and passenger perception.

[0035] S2. Collect static environmental parameters of the track surrounding environment and dynamic environmental parameters of vehicle operation, establish a multivariate regression model of environmental parameters and vehicle vibration, and quantify the influence weight of each environmental parameter on vibration.

[0036] I. Track Environmental Parameter Acquisition Static environmental parameters (sampling cycle 1 time / week): Track foundation settlement: measured using a total station, accuracy ±0.1mm, measurement point spacing 50m; Tunnel structure deformation: collected by shield segment joint displacement gauge, including radial and circumferential displacement, accuracy ±0.05mm; Rail profile: measured using a laser profiler, sampling interval 0.5mm, calculating parameters such as wheel-rail contact angle and equivalent taper; Track bed stiffness: measured using a falling weight deflectometer, with a measuring point every 10m, recording dynamic stiffness values ​​(kN / m), etc.

[0037] Dynamic environmental parameters (sampling cycle 1 time / minute): Train speed: calculated by onboard GPS and wheel diameter, accuracy ±1km / h; Passenger capacity: counted based on infrared sensors at the doors, error ≤5%; Track curve parameters: including curve radius, superelevation, transition curve length and other line parameters; Ambient temperature: temperature sensors are installed in the tunnel, measurement range -10~50℃, accuracy ±0.5℃, etc.

[0038] II. Construction of Multivariate Regression Model Input variables: 12 key environmental parameters were selected. This includes settlement, tunnel deformation, track bed stiffness, and operating speed; Output variable: Composite vibration index ( CVI ); Model Form: A multivariate nonlinear regression model is used, considering the interaction between parameters. The formula is as follows: ,in, For constant terms, The coefficients are linear. The coefficient of the interaction term. The error term (following a normal distribution) ); Model training: The least squares method is used to estimate the parameters, and the stepwise regression method is used to screen significant variables (P value < 0.05). The R² is used to evaluate the fit (R² ≥ 0.85 is required).

[0039] III. Calculation of Influence Weights The influence weights of each parameter are calculated based on the standardized regression coefficients, using the following formula: ,in These are standardized regression coefficients, reflecting the effect of parameters on regression. CVI The relative intensity of the influence.

[0040] S3. Based on the main environmental influencing factors identified by the aforementioned influence weights, the vibration transmission characteristics are simulated using a car-track rigid-flexible coupling model, and energy flow analysis is performed using a car-bogie finite element model to determine the key vibration transmission points that cause abnormal shaking in order to locate the fault source.

[0041] I. Simulation of the rigid-flexible coupling model of the vehicle body and track Model building: such as Figure 3 As shown, a rigid-flexible coupled vehicle-track model is built in Simpack software, which includes the following subsystems: Vehicle subsystems: flexible car body, bogie, flexible wheelsets, suspension system including primary and secondary springs and damping; Track subsystem: flexible steel rails, sleepers, and ballast bed, taking into account the elastic deformation of the track; Wheel-rail contact model: Hertz nonlinear contact is used to calculate the normal force, and Kalker Fastsim algorithm is used to calculate the tangential force.

[0042] Model parameters: Input actual operating vehicle parameters, track parameters (such as rail type 60kg / m, rail profile, etc.) and environmental parameters (such as track bed stiffness, etc.).

[0043] Simulation Analysis: Dynamic simulations are performed for different vibration levels, and the vibration transfer function (vibration amplification factor from wheelset to vehicle body) is output. The formula is: ,in, This represents the acceleration amplitude of the corresponding wheelset. This represents the acceleration amplitude of the corresponding wheelset.

[0044] II. Finite Element Vibration Transmission Path Analysis Model creation: A finite element model of the car body-bogie was built in Abaqus with a mesh size of 10-20mm and an element type of C3D8R, including details such as welding and bolt connections; Boundary conditions: Constrain the wheelset degrees of freedom and apply measured wheel-rail excitation forces; Vibration response analysis: Perform transient dynamic analysis, calculate the vibration acceleration of each node of the vehicle body in the 0-200Hz frequency band, and generate vibration cloud map; Key transmission point identification: The vibration energy transfer rate of each connection node is calculated using energy flow analysis. The formula is as follows: ,in, For the first k The vibrational energy transmitted by each node, where K is the total number of nodes, is selected. The node is used as the key transmission point.

[0045] III. Fault Source Location The main environmental influencing factors were identified by combining a multivariate regression model, the wheel-rail vibration transmission characteristics were analyzed by rigid-flexible coupling simulation, the key transmission points were located by finite element analysis, and a three-dimensional positioning model of "factor-path-node" was constructed. Fault type judgment: Based on the location results, match the fault knowledge base (containing 32 typical fault types) and output the probability distribution of fault causes. For example, if the track bed stiffness is reduced (weight 0.35) and the vibration in the 20-30Hz frequency band is significant, it is judged as insufficient track bed elasticity; if the wheel-rail contact angle is abnormal (deviation > 1°) and the lateral vibration energy accounts for > 60%, it is judged as polygonal wear of the wheelset.

[0046] S4. Based on the located fault source, generate temporary protection measures and a repair plan. The temporary protection measures are speed adjustment curves generated by combining genetic algorithm optimization, and the repair plan is a personalized work instruction generated based on the fault-repair process mapping library.

[0047] I. Setting Constraints Safety constraints: maximum deceleration ≤ 0.5 m / s² (to avoid passenger discomfort), minimum operating speed ≥ 20 km / h (straight section) or 10 km / h (curved section); Time constraints: Must meet the train timetable schedule, one-way delay time ≤ 5 minutes; Route constraints: Considering route conditions such as curves and slopes, the speed limit varies for different road sections (e.g., 40 km / h for small radius curves).

[0048] II. Construction of the Objective Function With "safety-efficiency-energy consumption" as the core of multi-objective optimization, the objective function is: ,in: Extend the runtime by (minutes). ; The number of vibration exceeding the standard ( CVI (Number of times >0.5 m / s²) The energy consumption increment (kWh) is proportional to the square of the speed curve; Weighting coefficients ( ), dynamically adjust according to the level of jitter (e.g., when jitter level 5). (Prioritize safety).

[0049] III. Genetic Algorithm Optimization Process Encoding method: Real number encoding is used to discretize the velocity curve into 10 control points (one point every 100m), and the chromosome length is 10; Initial population: 50 feasible solutions (velocity curves that satisfy the constraints) are randomly generated; Fitness function: (The smaller the objective function value, the higher the fitness). Selection operation: A roulette wheel selection method is used, and the selection probability is directly proportional to the fitness. Crossover operation: Arithmetic crossover is used, with a crossover probability of 0.8. The formula is as follows: , Mutation operation: Gaussian mutation is used with a mutation probability of 0.05. The formula is as follows: in, The standard deviation of the variation (decreases with the number of iterations, with an initial value of 5 km / h).

[0050] Termination criteria: No significant improvement in fitness after 30 iterations or 5 consecutive generations (change <1%).

[0051] IV. Generation of Speed ​​Adjustment Scheme Output the optimal speed curve: with distance as the x-axis and speed as the y-axis, including specific parameters for the acceleration segment, constant speed segment, and deceleration segment; Supporting measures: Generate driver operation instructions, such as "Start decelerating at K2+300, and reduce speed to 40 km / h within 500m". The operational guidelines have been refined: speed control points for different road sections have been clearly defined, such as "at K5+200 with a curve radius R=300m, the maximum speed shall not exceed 35km / h" and "when passing through the track bed settlement area (K8+100 to K8+500), maintain a constant speed of 25km / h". Vibration monitoring points at key locations have also been marked to remind drivers to pay attention to real-time vibration feedback.

[0052] Vehicle-to-ground coordination: The optimized speed curve is synchronized to the dispatch center via the vehicle-to-ground wireless communication system (transmission delay ≤100ms). The dispatch center performs a secondary confirmation based on real-time traffic conditions (such as the spacing between subsequent trains and passenger flow at platforms). Once confirmed, the information is fed back to the onboard system for execution. If a conflict exists, a rapid re-optimization mechanism is triggered, generating an alternative solution within 10 seconds.

[0053] V. Performance Evaluation Indicators Vibration control effect: calculated after speed adjustment CVI The reduction rate is calculated using the following formula: It requires a vibration level of 4 or higher. Operational efficiency loss: The ratio of one-way delay time to planned time, calculated using the following formula: ,Require This ensures that the impact on the overall network scheduling is minimized.

[0054] Energy consumption change rate: Comparing the energy consumption difference before and after optimization, the formula is: ,Require To avoid a surge in energy consumption due to excessive deceleration.

[0055] VI. Fault Types and Repair Please see Figure 4 The diagram shown is a logical diagram of the construction of a fault type and repair process mapping library proposed in this application embodiment.

[0056] Fault Classification System: Faults related to abnormal vehicle vibration are classified into 6 major categories and 32 subcategories, specifically including: Wheel-rail contact faults (8 types): including wheelset polygonal wear, rail corrugation, wheel-rail profile mismatch, etc.; Suspension system faults (7 types): including primary spring stiffness decay, secondary damping failure, air spring leakage, etc.; Track structure faults (6 types): including uneven track bed stiffness, track geometric deviation, loose rail fasteners, etc.; Vehicle body structure faults (5 types): including vehicle body modal resonance, loose connecting bolts, weld fatigue cracking, etc.; Environmental impact faults (4 types): including tunnel structure deformation, geological settlement, temperature stress, etc.; Other faults (2 types): including signal interference, abnormal software parameters, etc.

[0057] Standardize the repair process: Develop a standard repair procedure (SOP) for each type of fault, including: a list of required tools (such as torque wrenches, laser alignment instruments, etc.), key process parameters, quality inspection standards, etc.

[0058] VII. Spare parts inventory and vehicle model matching Spare parts database integration: Real-time communication with the Metro Group's inventory management system (WMS) to obtain information such as spare parts model, quantity, and storage location, with a data update frequency of 1 minute / time.

[0059] Vehicle compatibility verification: Establish a matching matrix between vehicle models and spare parts models to avoid mismatches.

[0060] Inventory warning mechanism: When the inventory of key spare parts is lower than the safety threshold (enough for 2 maintenance needs), a procurement warning is automatically triggered, a procurement list is generated and pushed to the materials management department.

[0061] VIII. Generation of Personalized Work Instructions 3D visualization: Generate a 3D schematic diagram of the fault location based on the BIM model, and mark the spatial location of the component to be repaired, surrounding related components, and disassembly sequence.

[0062] Step-by-step process: The repair process is broken down into several steps, each of which includes operating instructions, diagrams, and precautions.

[0063] Work and resource estimation: Based on historical maintenance data, estimate the work time and required personnel configuration for each operation step (e.g., 2 mechanics + 1 electrician), and generate a maintenance plan Gantt chart.

[0064] Version management: The work instructions are automatically linked to the vehicle number, the date of the failure, and the information of the repair personnel, forming a traceable version record for easy quality traceability.

[0065] IX. Scheme Review and Optimization Expert database access: For complex faults (such as vehicle body modal resonance), the system automatically invites more than 3 experts in relevant fields (vibration analysis, vehicle structure, etc.) to conduct online review, and the experts can mark modification opinions through the system.

[0066] Solution optimization: Adjust the repair process based on expert opinions, such as adding a "modal testing verification" step to ensure the feasibility of the solution. The optimized solution must be verified again by the system (meeting safety, quality, and efficiency requirements) before it can be output.

[0067] In other embodiments, the method of the present invention may optionally include the following steps: S5. After implementing the above-mentioned repair plan, conduct a trial run, collect vibration data after repair and compare it with that before repair, and quantitatively evaluate the repair effect.

[0068] I. Trial Operation Plan Design Test condition settings: Simulate typical operating conditions in actual operation, including: Speed ​​conditions: 60km / h (inter-station cruising), 80km / h (inter-section high speed), 30km / h (deceleration upon entering a station); Load conditions: Unloaded (0 people), Fully loaded (100% of rated passenger capacity), Overloaded (120%). Line conditions: Straight section, curve section with R=300m, and gradient (3‰).

[0069] Data Acquisition Plan: During the trial operation, vibration signals, passenger feedback, and environmental parameters will be collected according to the standard in step S1, with a sampling duration of ≥2 hours per operating condition to ensure data representativeness.

[0070] II. Quantitative Indicators of Repair Effectiveness Vibration improvement rate: Calculate the comprehensive vibration index before and after repair ( CVI The rate of change of ) is given by the formula: Requirements for vibration levels 3 and above. Level 1-2 jitter, Improvement in comfort level: Comparison of group comfort index before and after restoration ( GCI The difference between the two is calculated using the following formula: ,Require This means that the comfort level should be increased by at least one level.

[0071] Key component status: Inspect and repair the performance parameters of components, such as: Suspension system: primary spring stiffness deviation ≤3%, damping coefficient deviation ≤5%; Wheelset: wheel diameter difference ≤0.3mm, polygonal wear amplitude ≤0.05mm; Track: track bed stiffness uniformity (standard deviation ≤10%), track direction deviation ≤1mm / 30m.

[0072] III. Data Comparison and Report Generation Multi-dimensional comparison: Generating vibration spectrum comparison charts before and after repair. CVI Time series variation curves GCI The distribution histogram visually demonstrates the improvement effect. For example, in the 20-30Hz frequency band, the proportion of vibration energy decreased from 45% before the repair to 15%, indicating that the problem of insufficient track bed stiffness has been solved.

[0073] Verification Report: Automatically generates a repair effectiveness verification report containing the following: basic fault information (vehicle number, fault type, repair date), comparison of indicators before and after repair, assessment of compliance, remaining issues and improvement suggestions, etc. IV. Secondary Repair Trigger Mechanism If the repair effect does not meet the standard (e.g.) The system automatically triggers a secondary repair process: the system first re-analyzes the cause of the fault (calling the correlation analysis module in step S3); then generates alternative repair solutions; and increases the priority, including the vehicle in the emergency repair plan, and processes it again within 48 hours.

[0074] In other embodiments, the method of the present invention may optionally include the following steps: S6. Establish a long-term tracking mechanism for repaired vehicles and build a cross-line big data platform for data sharing and common problem analysis.

[0075] I. Long-term tracking indicator system Vibration trend monitoring: Vibration data are collected once a week and calculated. CVI Plot a trend curve based on the weekly average over three consecutive weeks. CVI When the increase is ≥10%, a warning will be issued (potential failure risk).

[0076] Component life prediction: Based on the vibration history of key components, a fatigue cumulative damage model is used to predict the remaining life, as shown in the formula: Where D is the cumulative damage value ( (Needs to be replaced) The number of cycles at a given stress level. The fatigue life at this stress level is obtained through the SN curve.

[0077] Environmental adaptability assessment: Track vibration data under different seasons and climatic conditions, such as the impact of track bed stiffness changes on vibration during the rainy season, and establish an environmental adaptability evaluation model.

[0078] II. Construction of a cross-line big data platform Data standardization: Establish unified data acquisition standards (sensor model, sampling frequency, parameter definition) to ensure the comparability of data from different lines. For example, the "lateral stability index" is uniformly defined as "the weighted root mean square value of acceleration in the 3Hz frequency band".

[0079] Database architecture: Utilizing a distributed database (such as Hadoop) to store multi-line data, including: Basic information database: route parameters (length, curve percentage, etc.), vehicle parameters (model, service life, etc.); Vibration database: CVI Spectrum characteristics, passenger feedback, etc.; Maintenance database: fault types, repair solutions, effectiveness evaluations, etc.

[0080] Data sharing mechanism: Set data access permissions (e.g., group-level users can view data for the entire line, while line-level users can only view data for their own line), and support cross-line data query and comparative analysis.

[0081] III. Common Problems and Countermeasures Cluster analysis: The K-means algorithm is used to cluster the fault data of multiple lines to identify common problems (such as "all lines crossing soft soil layers have vibrations caused by track bed settlement").

[0082] Root cause analysis: For common problems, conduct special studies (such as the matching analysis of geological conditions and track structure) and propose systematic solutions (such as "using floating slab track in soft soil layers").

[0083] Standard optimization: Based on the analysis results, vehicle maintenance standards and track design specifications are revised to reduce the occurrence of failures at the source. For example, the "wheelset turning cycle" is optimized from "80,000 kilometers" to "dynamically adjusted according to the wear trend of polygons".

[0084] IV. Forward-looking early warning model Input variables: integrate long-term tracking data (vibration trends, component status), environmental parameters (season, geology), and operational data (passenger capacity, speed); Model Construction: An LSTM neural network is used to predict the jitter level for the next 3 months. The formula is as follows: ,in, This is a predicted jitter level for 3 months later. Historical jitter level, Here, f represents the historical environment and operational parameters, and f is the LSTM network mapping function.

[0085] Early warning response: When the predicted jitter level is ≥3, preventive measures (such as planned maintenance and track pre-adjustment) should be formulated in advance to avoid failure.

[0086] Example 2 Please see Figure 5 The diagram shown is a structural schematic of a subway vehicle abnormal vibration repair system according to the second embodiment of this application. The system includes the following key modules: The vehicle body abnormal vibration classification module 100 is used to collect vehicle body vibration signals and integrate passenger comfort feedback information to establish a vehicle body abnormal vibration classification system that correlates objective vibration indicators with subjective passenger feelings. The track environment parameter quantification module 200 is used to collect static environmental parameters of the track surrounding environment and dynamic environmental parameters of vehicle operation, establish a multivariate regression model of environmental parameters and vehicle vibration, and quantify the influence weight of each environmental parameter on vibration. The environment-vibration correlation analysis fault source module 300 is used to identify the main environmental influencing factors based on the influence weights, simulate the vibration transmission characteristics through the car body-track rigid-flexible coupling model, and perform energy flow analysis in combination with the car body-bogie finite element model to determine the key vibration transmission points that cause abnormal vibration in order to locate the fault source. The intelligent decision-making and repair strategy generation module 400 is used to generate temporary protection measures and a repair plan for inclusion in the database based on the located fault source. The temporary protection measures are speed adjustment curves generated by combining genetic algorithms, and the repair plan for inclusion in the database is a personalized work instruction generated based on the fault-repair process mapping library.

[0087] Optionally, the system also includes the following key modules: The repair execution and closed-loop verification module 500 is used to perform a trial run after the repair plan is executed, collect vibration data after repair and compare it with that before repair, and quantitatively evaluate the repair effect. The Long-Term Tracking and Cross-Line Big Data Module 600 is used to establish a long-term tracking mechanism for repaired vehicles and build a cross-line big data platform for data sharing and common problem analysis.

[0088] The subway vehicle abnormal vibration repair system in this application embodiment can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, mobile electronic devices can be mobile phones, tablets, laptops, PDAs, in-vehicle electronic devices, wearable devices, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), etc., while non-mobile electronic devices can be servers, network attached storage (NAS), personal computers (PCs), etc., and this application embodiment does not specifically limit the scope.

[0089] The subway vehicle abnormal vibration repair system in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit it.

[0090] This application provides a subway vehicle abnormal vibration repair system that can achieve... Figure 1 The various processes implemented in the method embodiment of a subway vehicle abnormal vibration repair method are not described in detail here to avoid repetition.

[0091] Optionally, this application embodiment also provides an electronic device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, they implement the various processes of the above-described embodiment of a method for repairing abnormal shaking in subway vehicles and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0092] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described embodiment of a method for repairing abnormal shaking in subway vehicles and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0093] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0094] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0095] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, 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 is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0096] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for repairing abnormal vibrations in subway vehicles, characterized in that, Includes the following steps: S1. Collect vehicle vibration signals and integrate passenger comfort feedback information to establish a vehicle abnormal vibration classification system that correlates objective vibration indicators with subjective passenger feelings. S2. Collect static environmental parameters of the track surrounding environment and dynamic environmental parameters of vehicle operation, establish a multivariate regression model of environmental parameters and vehicle vibration, and quantify the influence weight of each environmental parameter on vibration. S3. Based on the main environmental influencing factors identified by the aforementioned influence weights, the vibration transmission characteristics are simulated using a car-track rigid-flexible coupling model, and energy flow analysis is performed using a car-bogie finite element model to determine the key vibration transmission points that cause abnormal shaking in order to locate the fault source. S4. Based on the located fault source, generate temporary protection measures and a repair plan. The temporary protection measures are speed adjustment curves generated by combining genetic algorithm optimization, and the repair plan is a personalized work instruction generated based on the fault-repair process mapping library.

2. The method for repairing abnormal vibrations in subway vehicles according to claim 1, characterized in that, Step S1 specifically includes: ICP-type triaxial acceleration sensors are installed on the top front of the driver's cab, on both sides of the middle of the car body, on the front and rear side walls of the passenger compartment, and on the bottom of the car body above the bogie. The acquired raw vibration signal was subjected to high-pass filtering to remove the DC component, Butterworth low-pass filtering to suppress high-frequency noise, and the 3σ criterion was used to eliminate impulse noise. The root mean square value of acceleration (RMS) and peak factor (CF) of the preprocessed signal are calculated, and the main vibration frequency components are identified. The comprehensive vibration index (CVI) is constructed by weighted summation as an objective vibration indicator.

3. The method for repairing abnormal vibrations in subway vehicles according to claim 2, characterized in that, Step S1 also specifically includes: Passengers are given a 1-5 point rating on five core indicators: vibration intensity, vibration frequency, duration, standing stability, and overall comfort, through an in-vehicle application or QR code. The individual comfort index (ICI) is constructed by weighting and summing the scores of the five indicators in a single feedback. The Individual Comfort Index (ICI) of all feedback from the same train trip is statistically analyzed to construct the Group Comfort Index (GCI). Using the Comprehensive Vibration Index (CVI) and the Group Comfort Index (GCI) as two dimensions, a shaking classification system with five levels—slight, mild, moderate, severe, and extremely severe—is established.

4. A method for repairing abnormal vibrations in subway vehicles according to claim 1 or 3, characterized in that, Steps S2 and S3 specifically include: The collected static environmental parameters include: track foundation settlement measured by a total station, tunnel structure deformation collected by a shield segment joint displacement gauge, rail profile measured by a laser profiler, and track bed stiffness measured by a falling weight deflectometer. The collected dynamic environmental parameters include: train speed calculated using onboard GPS and wheel diameter, passenger capacity based on door infrared sensor counts, track curve parameters, and tunnel temperature. The car body-track rigid-flexible coupling model was established using Simpack software. The vehicle subsystem includes a flexible car body, rigid bogie, flexible wheelset and suspension system. The track subsystem includes flexible rails, sleepers and track bed. The Hertz nonlinear contact calculation was used to calculate the normal force of the wheel-rail contact, and the Kalker Fastsim algorithm was used to calculate the tangential force. The car body-bogie finite element model was constructed using Abaqus software. Transient dynamic analysis was performed by applying measured wheel-rail excitation forces. The vibration energy transfer rate of each connection node was calculated using the energy flow analysis method, and the node with the highest vibration energy transfer rate was identified as the key vibration transmission point.

5. The method for repairing abnormal vibrations in subway vehicles according to claim 1, characterized in that, In step S4, the generation of the speed adjustment curve includes: Construct a multi-objective optimization function with the objectives of extending runtime, the number of vibration exceedances, and the energy consumption increment; The constraints include: safety constraints, wherein the maximum deceleration acceleration is not greater than 0.5 m / s², the minimum operating speed on straight sections is not less than 20 km / h, and the minimum operating speed on curved sections is not less than 10 km / h; time constraints, wherein the one-way delay time is not greater than 5 minutes; and route constraints, which include speed limits under different curve and gradient conditions. The multi-objective optimization function is solved using a genetic algorithm under the constraints to obtain the optimal speed adjustment curve, which is then pushed to the dispatch center via the vehicle-to-ground communication system.

6. The method for repairing abnormal vibrations in subway vehicles according to claim 1, characterized in that, Step S4 specifically includes: Establish a fault-repair process mapping library that classifies faults into 6 major categories and 32 subcategories, namely wheel-rail contact, suspension system, track structure, car body structure, environmental impact, and other. The library contains a standard repair process for each type of fault, which includes a list of required tools, key process parameters, and quality inspection standards. The personalized work instruction manual is generated based on the BIM model. Its content includes a three-dimensional schematic diagram of the fault location, step-by-step operation instructions, precautions, and time and resource estimates.

7. A method for repairing abnormal vibrations in subway vehicles according to claim 1 or 6, characterized in that, The method further includes the following steps: S5. After implementing the aforementioned repair plan, conduct a trial run, collect vibration data after repair and compare it with that before repair, and quantitatively evaluate the repair effect. The indicators for quantitatively evaluating the repair effect include: vibration improvement rate, where the vibration improvement rate for level 3 and above vibration is not less than 70%, and for level 1-2 vibration is not less than 50%; comfort improvement rate, where the difference in the Group Comfort Index (GCI) before and after repair is not less than 1.0; and the condition of key components, where the primary spring stiffness deviation is not greater than 3%, and the wheel set diameter difference is not greater than 0.3mm. S6. Establish a long-term tracking mechanism for repaired vehicles and build a cross-line big data platform for data sharing and common problem analysis. The construction of the cross-line big data platform includes: collecting vibration data weekly to calculate the weekly average of the Comprehensive Vibration Index (CVI); issuing an early warning when the CVI increases by at least 10% for three consecutive weeks; predicting the remaining lifespan of components based on the vibration history of key components using a fatigue cumulative damage model; using the K-means algorithm to cluster fault data from multiple lines to identify common problems and revise vehicle maintenance standards; and using an LSTM neural network to build a forward-looking early warning model, inputting long-term tracking data, environmental parameters, and operational data to predict the vibration level for the next three months. When the predicted vibration level reaches level 3 or above, preventative maintenance measures are formulated in advance.

8. A subway vehicle abnormal vibration repair system, characterized in that, include: The vehicle body abnormal vibration classification module is used to collect vehicle body vibration signals and integrate passenger comfort feedback information to establish a vehicle body abnormal vibration classification system that correlates objective vibration indicators with subjective passenger feelings. The track environment parameter quantification module is used to collect static environmental parameters of the track surrounding environment and dynamic environmental parameters of vehicle operation, establish a multivariate regression model of environmental parameters and vehicle vibration, and quantify the influence weight of each environmental parameter on vibration. The environment-vibration correlation analysis fault source module is used to identify the main environmental influencing factors based on the influence weights, simulate the vibration transmission characteristics through the car body-track rigid-flexible coupling model, and perform energy flow analysis in combination with the car body-bogie finite element model to determine the key vibration transmission points that cause abnormal vibration in order to locate the fault source. The intelligent decision-making and repair strategy generation module is used to generate temporary protection measures and a repair plan for inclusion in the database based on the located fault source. The temporary protection measures are speed adjustment curves generated by combining genetic algorithms, and the repair plan for inclusion in the database is a personalized work instruction generated based on the fault-repair process mapping library.

9. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, they implement the steps of a subway vehicle abnormal vibration repair method as described in any one of claims 1-7.

10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions, which, when executed by a processor, implement the steps of a subway vehicle abnormal vibration repair method as described in any one of claims 1-7.

Images