Traction transmission system fault diagnosis and health prediction system and method thereof

Abstract

The invention discloses a traction transmission system fault diagnosis and health prediction system and method, and relates to the technical field of railway tracks, and the system comprises a data collection system, a track-vehicle system model, a traction transmission system model, and a Sichuan-Tibetan line train running gear instability evaluation algorithm. According to the traction transmission system fault diagnosis and health prediction system and the method thereof, a track-vehicle system model is established based on acquired Sichuan and Tibet railway and vehicle parameters, a traction transmission system model is further constructed, a real-time track-vehicle relationship is simulated with the assistance of multi-physical field coupling, and the fault diagnosis and health prediction of the traction transmission system are realized. The model effect is enhanced from five dimensions of a physical entity, a virtual entity, twinborn data, a connection rule and application service, so that a five-dimensional model which is more practical is obtained, the risk condition when a vehicle enters a next road section is calculated in advance by utilizing an instability judgment algorithm, multi-modal, multi-parameter and multi-disciplinary information is fused, and the reliability of the vehicle is improved. And powerful guarantee is provided for safe and efficient operation of Sichuan and Tibet railways by utilizing advanced technical means.

Description

Technical Field

[0001] The present invention relates to the technical field of railway tracks, and in particular to a traction drive system fault diagnosis and health prediction system and a method thereof. Background Art

[0002] The Sichuan-Tibet Railway is located in a plateau area with complex terrain, harsh climate and extreme natural conditions. The line is 1,956 kilometers long, with an average altitude of more than 4,000 meters and the highest point of 5,073 meters. The line passes through many high mountain tunnels, bridges and extra-large bridges, and the safety of the line is greatly challenged. Therefore, a systematic method is needed to monitor the safety of the Sichuan-Tibet Railway line and vehicles in real time and make predictions.

[0003] Therefore, in view of this, the existing structure and deficiencies are studied and improved, and a traction drive system fault diagnosis and health prediction system and method are proposed. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a traction drive system fault diagnosis and health prediction system and method thereof, which solves the problems raised in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a traction drive system fault diagnosis and health prediction system, the traction drive system fault diagnosis and health prediction system comprising:

[0006] A data collection system, wherein the data collection system is used to collect line parameters of the Sichuan-Tibet Railway and dynamic parameters of EMUs;

[0007] A track-vehicle system model, wherein the track-vehicle system model is used to establish a geometric model and a structural model of the track, a multi-body dynamics model of the vehicle, and a three-dimensional wheel-rail contact model to simulate the force and deformation between the wheel and rail;

[0008] A traction drive system model, wherein the traction drive system model is used to establish a motor model, a drive system model, and a traction control model;

[0009] Based on the above systems and models, multi-physics field coupling is realized, including force field coupling, i.e. mechanical coupling of track-vehicle system, electro-mechanical coupling, i.e. electro-mechanical coupling of traction motor and transmission system, and thermodynamic coupling, i.e. thermodynamic coupling of motor and transmission system;

[0010] A railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture, wherein the railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture is used to construct a 5-dimensional model based on physical entities, virtual entities, twin data, connection rules, and application services;

[0011] The Sichuan-Tibet line train running gear instability assessment algorithm is used to perform instability assessment on the vehicle in real time when it passes through any section of the road based on the railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture.

[0012] Furthermore, the Sichuan-Tibet Railway line parameters include line geometric parameters, track parameters, and environmental conditions, and the EMU dynamic parameters include traction motor characteristics, gear transmission characteristics, and traction control system parameters.

[0013] Furthermore, the geometric model of the track includes the spatial position and shape of the track, the structural model of the track includes the stiffness and damping characteristics of the track, and the multi-body dynamic model of the vehicle includes the body, bogie, wheelset, suspension system, and each part needs to consider its mass, inertia, stiffness and damping characteristics.

[0014] Furthermore, the motor model is an electromagnetic and mechanical model of the traction motor, which is used to simulate the dynamic characteristics and control characteristics of the motor. The transmission system model includes the stiffness, damping, and gear meshing error of the gear transmission system. The traction control model simulates the traction control algorithm, including speed control and torque control.

[0015] Furthermore, the physical entity includes:

[0016] Track system: track geometry, track structure, and track environment;

[0017] Vehicle system: car body, bogie, wheelset, braking system, suspension system;

[0018] Traction drive system: traction motor, converter, transmission device, control system;

[0019] Virtual entities include:

[0020] Simulation model: Build a high-precision simulation model based on physical entities to simulate system operation status and predict potential failures;

[0021] Digital twin model: digitizes physical entities and builds virtual entities that are synchronized with physical entities in real time to achieve status monitoring, fault diagnosis, and health prediction functions;

[0022] Twin data includes:

[0023] Real-time data: real-time operation data of tracks, vehicles, and traction drive systems obtained from sensors, monitoring equipment, etc.;

[0024] Historical data: system operation historical data, used for data analysis and model training;

[0025] Simulation data: Data generated by the operation of simulation models, used to verify models, optimize algorithms, and other connection rules and application services.

[0026] Furthermore, the connection rules include:

[0027] Data connection: establish data connection between physical entities and virtual entities to achieve real-time data synchronization;

[0028] Model connection: Establish connections between different systems and different disciplines to achieve information sharing and collaborative analysis;

[0029] Service connection: Establish a connection between the digital twin and application services to realize fault diagnosis and health prediction functions.

[0030] Furthermore, the application service includes:

[0031] Condition monitoring: real-time monitoring of the operating status of tracks, vehicles, and traction drive systems, and identification of abnormal conditions;

[0032] Fault diagnosis: Based on data analysis and model reasoning, diagnose the cause of system failure and provide fault location and solutions;

[0033] Health prediction: Based on historical data and model predictions, evaluate system health status, predict potential failures, and develop maintenance plans.

[0034] Furthermore, the formula for judging the instability of the running gear of the Sichuan-Tibet Line train is as follows:

[0035] R=f(V,A_h,θ,F_l,W,E);

[0036] Where, V: train speed;

[0037] A_h: lateral acceleration;

[0038] θ: train tilt angle;

[0039] F_l: wheel-rail lateral force;

[0040] W: vehicle weight;

[0041] E: Environmental factors;

[0042] R: instability risk score;

[0043] The function f is either linear or nonlinear.

[0044] Furthermore, the determination of the instability evaluation is as follows:

[0045] If R exceeds a predetermined threshold R_{threshold}, the train is considered to be at risk of instability.

[0046] A method for a traction drive system fault diagnosis and health prediction system is applied with the above-mentioned traction drive system fault diagnosis and health prediction system, and the method for a traction drive system fault diagnosis and health prediction system comprises the following steps:

[0047] Step 1: Use various sensors or parameter detection equipment to obtain the line parameters of the Sichuan-Tibet Railway and the dynamic parameters of the EMU in real time, and build a track-vehicle system model based on the various parameters obtained;

[0048] The data is dynamically compensated according to the real-time environmental parameters, including temperature and air pressure. The specific steps are as follows:

[0049] Conduct laboratory tests to obtain the response characteristics of the sensor at different temperatures and air pressures;

[0050] Build a mathematical model to describe the relationship between environmental parameters and sensor output;

[0051] The compensation algorithm is as follows:

[0052] For temperature compensation, use the following formula:

[0053] Compensated_Value=Raw_Value+K*(T-T_ref)

[0054] Among them, Compensated_Value is the compensated value;

[0055] Raw_Value is the raw measurement value;

[0056] K is the temperature coefficient, which is determined experimentally;

[0057] T is the current temperature;

[0058] T_ref is the reference temperature;

[0059] Multi-parameter influences such as the combined effects of temperature and air pressure are also taken into account using multivariable compensation models, for example:

[0060] Compensated_Value=Raw_Value+f(T,P)

[0061] Where f(T,P) is a function of temperature T and pressure P;

[0062] Step 2: Based on the track-vehicle system model, the traction drive system model is further constructed according to the parameters, and multi-physics field coupling is realized to simulate the real-time track-vehicle relationship. The model effect is enhanced from five dimensions: physical entity, virtual entity, twin data, connection rules, and application services, thereby obtaining a more realistic five-dimensional model.

[0063] Step 3: retrieve various parameters of the vehicle when it is about to pass through the section. The prediction of such parameters is obtained through reciprocating actual tests. The parameters are brought into the instability evaluation algorithm to calculate the instability risk of the vehicle on this section. When the result is greater than the preset threshold, it is considered that the train has the risk of instability, and the staff is automatically prompted to take corresponding treatment measures;

[0064] Step 4: Based on real-time monitoring of various vehicle parameters, data analysis and model reasoning, diagnose the cause of system failure, provide fault location and solutions, and evaluate the system health status based on historical data and model predictions, predict potential failures, and formulate maintenance plans.

[0065] The present invention provides a traction drive system fault diagnosis and health prediction system and method thereof, which has the following beneficial effects:

[0066] 1. The traction drive system fault diagnosis and health prediction system and its method use various sensors or parameter detection equipment to obtain the parameters of the Sichuan-Tibet Railway and vehicles in real time, establish a track-vehicle system model, and further construct a traction drive system model, and then use multi-physical field coupling to simulate the real-time track-vehicle relationship. The model effect is enhanced from five dimensions: physical entity, virtual entity, twin data, connection rules, and application services, thereby obtaining a more realistic five-dimensional model, and using an instability assessment algorithm to pre-calculate the risk of the vehicle entering the next section to predict the degree of risk in advance. This will integrate multi-modal, multi-parameter, and multi-disciplinary information, and use advanced technical means to provide strong guarantees for the safe and efficient operation of the Sichuan-Tibet Railway.

[0067] 2. The traction drive system fault diagnosis and health prediction system and method, by establishing a mathematical model to describe the relationship between environmental parameters and sensor output, dynamically compensates data based on real-time environmental parameters based on the response characteristics of the sensor at different temperatures and air pressures, which can significantly improve the accuracy of data in a changing environment and provide more reliable basic data for subsequent fault diagnosis and health prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a structural schematic diagram of the present invention;

[0069] Figure 2 It is a schematic diagram of the operation flow of the present invention. DETAILED DESCRIPTION

[0070] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0071] like Figure 1-Figure 2 As shown, the present invention provides a technical solution: a traction drive system fault diagnosis and health prediction system, the traction drive system fault diagnosis and health prediction system comprising:

[0072] A data collection system, wherein the data collection system is used to collect line parameters of the Sichuan-Tibet Railway and dynamic parameters of the EMU. The line parameters of the Sichuan-Tibet Railway include geometric parameters of the line (such as curvature, slope, route length, etc.), track parameters (such as track type, roadbed type, etc.), and environmental conditions (such as altitude, temperature, humidity, etc.). The dynamic parameters of the EMU include traction motor characteristics (such as power, torque, efficiency, etc.), gear transmission characteristics, traction control system parameters, etc.;

[0073] A track-vehicle system model, wherein the track-vehicle system model is used to establish a geometric model and a structural model of the track, a multi-body dynamics model of the vehicle, and a three-dimensional wheel-rail contact model to simulate the force and deformation between the wheel and the rail. The geometric model of the track includes the spatial position and shape of the track, and the structural model of the track includes the stiffness and damping characteristics of the track. The multi-body dynamics model of the vehicle includes the car body, bogie, wheelset, suspension system, etc., and each part needs to consider its mass, inertia, stiffness, damping and other characteristics;

[0074] A traction drive system model, wherein the traction drive system model is used to establish a motor model, a transmission system model, and a traction control model. The motor model is an electromagnetic and mechanical model of the traction motor, and is used to simulate the dynamic characteristics and control characteristics of the motor. The transmission system model includes the stiffness, damping, gear meshing error, etc. of the gear transmission system. The traction control model simulates the traction control algorithm, including speed control, torque control, etc.;

[0075] Based on the above systems and models, multi-physics field coupling is realized, including force field coupling, i.e. mechanical coupling of track-vehicle system, electro-mechanical coupling, i.e. electro-mechanical coupling of traction motor and transmission system, and thermodynamic coupling, i.e. thermodynamic coupling of motor and transmission system;

[0076] Railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture, the railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture is used to build a 5-dimensional model based on physical entities, virtual entities, twin data, connection rules, and application services to achieve multi-modal, multi-parameter, and multi-disciplinary digital twin imaging of the Sichuan-Tibet Railway-train system. Multi-modality: integrates multiple modal data such as sensor data, image data, and video data to build a more comprehensive system status perception;

[0077] Multi-parameter: Comprehensively consider multi-parameter information such as track geometry parameters, vehicle dynamics parameters, traction drive system electrical parameters, etc. to improve the accuracy of fault diagnosis and health prediction;

[0078] Multidisciplinary: Integrate multidisciplinary knowledge such as rail engineering, vehicle engineering, electrical engineering, and control engineering to build a more complete digital twin model;

[0079] Physical entities include:

[0080] Track system: track geometry, track structure, track environment, etc.;

[0081] Vehicle system: car body, bogie, wheelset, braking system, suspension system, etc.;

[0082] Traction drive system: traction motor, converter, transmission device, control system, etc.;

[0083] Virtual entities include:

[0084] Simulation model: Build a high-precision simulation model based on physical entities to simulate system operation status and predict potential failures;

[0085] Digital twin model: digitizes physical entities and builds virtual entities that are synchronized with physical entities in real time to achieve functions such as status monitoring, fault diagnosis, and health prediction;

[0086] Twin data includes:

[0087] Real-time data: real-time operation data of tracks, vehicles, and traction drive systems obtained from sensors, monitoring equipment, etc.;

[0088] Historical data: system operation historical data, used for data analysis, model training, etc.;

[0089] Simulation data: Data generated by the operation of simulation models, used to verify models, optimize algorithms, and other connection rules and application services.

[0090] The connection rules include:

[0091] Data connection: establish data connection between physical entities and virtual entities to achieve real-time data synchronization;

[0092] Model connection: Establish connections between different systems and different disciplines to achieve information sharing and collaborative analysis;

[0093] Service connection: Establish a connection between the digital twin and application services to realize functions such as fault diagnosis and health prediction.

[0094] Application services include:

[0095] Condition monitoring: real-time monitoring of the operating status of tracks, vehicles, and traction drive systems, and identification of abnormal conditions;

[0096] Fault diagnosis: Based on data analysis and model reasoning, diagnose the cause of system failure and provide fault location and solutions;

[0097] Health prediction: Based on historical data and model prediction, evaluate system health status, predict potential failures, and develop maintenance plans;

[0098] Through real-time monitoring and fault diagnosis, potential faults can be discovered and handled in a timely manner to ensure safe railway operation;

[0099] Sichuan-Tibet line train running gear instability evaluation algorithm, the Sichuan-Tibet line train running gear instability evaluation algorithm is used to perform instability evaluation of the vehicle passing through any section in real time according to the railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture. The Sichuan-Tibet line train running gear instability evaluation algorithm formula is as follows:

[0100] R=f(V,A_h,θ,F_l,W,E);

[0101] Where, V: train speed;

[0102] A_h: lateral acceleration;

[0103] θ: train tilt angle;

[0104] F_l: wheel-rail lateral force;

[0105] W: vehicle weight;

[0106] E: Environmental factors;

[0107] R: instability risk score;

[0108] The function f is either linear or nonlinear.

[0109] The determination of the instability evaluation is as follows:

[0110] If R exceeds a certain predetermined threshold R_{threshold}, the train is considered to be at risk of instability;

[0111] Based on the above description, the data collection system is connected to the track-vehicle system model, the track-vehicle system model is connected to the traction drive system model, and the traction drive system model is connected to the railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture;

[0112] The digital twin architecture for fault diagnosis and health prediction of railway track-vehicle-traction drive system is connected with the instability assessment algorithm of the running gear of Sichuan-Tibet Line trains.

[0113] The present invention utilizes various sensors or parameter detection equipment to obtain the parameters of the Sichuan-Tibet Railway and vehicles in real time, establishes a track-vehicle system model, and further constructs a traction transmission system model, and then uses multi-physical field coupling to simulate the real-time track-vehicle relationship. The model effect is enhanced from five dimensions: physical entity, virtual entity, twin data, connection rules, and application services, thereby obtaining a more practical five-dimensional model, and using an instability assessment algorithm to pre-calculate the risk of the vehicle when entering the next section, so as to predict the degree of risk in advance. In this way, multi-modal, multi-parameter, and multi-disciplinary information are integrated, and advanced technical means are used to provide strong guarantees for the safe and efficient operation of the Sichuan-Tibet Railway.

[0114] A method for a traction drive system fault diagnosis and health prediction system is applied with the above-mentioned traction drive system fault diagnosis and health prediction system, and the method for a traction drive system fault diagnosis and health prediction system comprises the following steps:

[0115] Step 1: Use various sensors or parameter detection equipment to obtain the line parameters of the Sichuan-Tibet Railway and the dynamic parameters of the EMU in real time, and build a track-vehicle system model based on the various parameters obtained;

[0116] The data is dynamically compensated according to the real-time environmental parameters, including temperature and air pressure. The specific steps are as follows:

[0117] Conduct laboratory tests to obtain the response characteristics of the sensor at different temperatures and air pressures;

[0118] Build a mathematical model to describe the relationship between environmental parameters and sensor output;

[0119] The compensation algorithm is as follows:

[0120] For temperature compensation, use the following formula:

[0121] Compensated_Value=Raw_Value+K*(T-T_ref)

[0122] Among them, Compensated_Value is the compensated value;

[0123] Raw_Value is the raw measurement value;

[0124] K is the temperature coefficient, which is determined experimentally;

[0125] T is the current temperature;

[0126] T_ref is the reference temperature;

[0127] Multi-parameter influences such as the combined effects of temperature and air pressure are also taken into account using multivariable compensation models, such as:

[0128] Compensated_Value=Raw_Value+f(T,P)

[0129] Where f(T,P) is a function of temperature T and pressure P;

[0130] Step 2: Based on the track-vehicle system model, the traction drive system model is further constructed according to the parameters, and multi-physics field coupling is realized to simulate the real-time track-vehicle relationship. The model effect is enhanced from five dimensions: physical entity, virtual entity, twin data, connection rules, and application services, thereby obtaining a more realistic five-dimensional model.

[0131] Step 3: retrieve various parameters of the vehicle when it is about to pass through the section. The prediction of such parameters is obtained through reciprocating actual tests. The parameters are brought into the instability evaluation algorithm to calculate the instability risk of the vehicle on this section. When the result is greater than the preset threshold, it is considered that the train has the risk of instability, and the staff is automatically prompted to take corresponding treatment measures;

[0132] Step 4: Based on real-time monitoring of various vehicle parameters, data analysis and model reasoning, diagnose the cause of system failure, provide fault location and solutions, and evaluate the system health status based on historical data and model predictions, predict potential failures, and formulate maintenance plans.

[0133] Based on the above description, by establishing a mathematical model to describe the relationship between environmental parameters and sensor output, a dynamic compensation method based on environmental real-time parameters is performed based on the response characteristics of the sensor at different temperatures and air pressures. This can significantly improve the accuracy of data in a changing environment and provide more reliable basic data for subsequent fault diagnosis and health prediction.

[0134] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A traction drive system fault diagnosis and health prediction system, characterized in that: The traction drive system fault diagnosis and health prediction system comprises: A data collection system, wherein the data collection system is used to collect line parameters of the Sichuan-Tibet Railway and dynamic parameters of EMUs; A track-vehicle system model, wherein the track-vehicle system model is used to establish a geometric model and a structural model of the track, a multi-body dynamics model of the vehicle, and a three-dimensional wheel-rail contact model to simulate the force and deformation between the wheel and rail; A traction drive system model, wherein the traction drive system model is used to establish a motor model, a drive system model, and a traction control model; Based on the above systems and models, multi-physics field coupling is realized, including force field coupling, i.e. mechanical coupling of track-vehicle system, electro-mechanical coupling, i.e. electro-mechanical coupling of traction motor and transmission system, and thermodynamic coupling, i.e. thermodynamic coupling of motor and transmission system; A railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture, wherein the railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture is used to construct a 5-dimensional model based on physical entities, virtual entities, twin data, connection rules, and application services; The Sichuan-Tibet line train running gear instability assessment algorithm is used to perform instability assessment on the vehicle in real time when it passes through any section of the road based on the railway track-vehicle-traction drive system fault diagnosis and health prediction digital twin architecture.

2. A traction drive system fault diagnosis and health prediction system according to claim 1, characterized in that: The Sichuan-Tibet Railway line parameters include line geometric parameters, track parameters, and environmental conditions; the EMU dynamic parameters include traction motor characteristics, gear transmission characteristics, and traction control system parameters.

3. A traction drive system fault diagnosis and health prediction system according to claim 1, characterized in that: The geometric model of the track includes the spatial position and shape of the track, the structural model of the track includes the stiffness and damping characteristics of the track, and the multi-body dynamics model of the vehicle includes the body, bogie, wheelset, suspension system, and each part needs to consider its mass, inertia, stiffness and damping characteristics.

4. A traction drive system fault diagnosis and health prediction system according to claim 1, characterized in that: The motor model is an electromagnetic and mechanical model of the traction motor, which is used to simulate the dynamic characteristics and control characteristics of the motor. The transmission system model includes the stiffness, damping, and gear meshing error of the gear transmission system. The traction control model simulates the traction control algorithm, including speed control and torque control.

5. A traction drive system fault diagnosis and health prediction system according to claim 1, characterized in that: The physical entity comprises: Track system: track geometry, track structure, and track environment; Vehicle system: car body, bogie, wheelset, braking system, suspension system; Traction drive system: traction motor, converter, transmission device, control system; Virtual entities include: Simulation model: Build a high-precision simulation model based on physical entities to simulate system operation status and predict potential failures; Digital twin model: digitizes physical entities and builds virtual entities that are synchronized with physical entities in real time to achieve status monitoring, fault diagnosis, and health prediction functions; Twin data includes: Real-time data: real-time operation data of tracks, vehicles, and traction drive systems obtained from sensors, monitoring equipment, etc.; Historical data: system operation historical data, used for data analysis and model training; Simulation data: Data generated by the operation of simulation models, used to verify models, optimize algorithms, and other connection rules and application services.

6. A traction drive system fault diagnosis and health prediction system according to claim 1, characterized in that: The connection rules include: Data connection: establish data connection between physical entities and virtual entities to achieve real-time data synchronization; Model connection: Establish connections between different systems and different disciplines to achieve information sharing and collaborative analysis; Service connection: Establish a connection between the digital twin and application services to realize fault diagnosis and health prediction functions.

7. A traction drive system fault diagnosis and health prediction system according to claim 1, characterized in that: The application services include: Condition monitoring: real-time monitoring of the operating status of tracks, vehicles, and traction drive systems, and identification of abnormal conditions; Fault diagnosis: Based on data analysis and model reasoning, diagnose the cause of system failure and provide fault location and solutions; Health prediction: Based on historical data and model predictions, evaluate system health status, predict potential failures, and develop maintenance plans.

8. A traction drive system fault diagnosis and health prediction system according to claim 1, characterized in that: The formula for judging the instability of the running gear of the Sichuan-Tibet Line train is as follows: R=f(V,A_h,θ,F_l,W,E); Where, V: train speed; A_h: lateral acceleration; θ: train tilt angle; F_l: wheel-rail lateral force; W: vehicle weight; E: Environmental factors; R: instability risk score; The function f is either linear or nonlinear.

9. A traction drive system fault diagnosis and health prediction system according to claim 8, characterized in that: The determination of the instability evaluation is as follows: If R exceeds a predetermined threshold R_{threshold}, the train is considered to be at risk of instability.

10. A method of a traction drive system fault diagnosis and health prediction system, which is applied with a traction drive system fault diagnosis and health prediction system according to any one of claims 1 to 9, characterized in that: The method of the traction drive system fault diagnosis and health prediction system comprises the following steps: Step 1: Use various sensors or parameter detection equipment to obtain the line parameters of the Sichuan-Tibet Railway and the dynamic parameters of the EMU in real time, and build a track-vehicle system model based on the various parameters obtained; The data is dynamically compensated according to the real-time environmental parameters, including temperature and air pressure. The specific steps are as follows: Conduct laboratory tests to obtain the response characteristics of the sensor at different temperatures and air pressures; Build a mathematical model to describe the relationship between environmental parameters and sensor output; The compensation algorithm is as follows: For temperature compensation, use the following formula: Compensated_Value=Raw_Value+K*(T-T_ref) Among them, Compensated_Value is the compensated value; Raw_Value is the raw measurement value; K is the temperature coefficient, which is determined experimentally; T is the current temperature; T_ref is the reference temperature; Step 2: Based on the track-vehicle system model, the traction drive system model is further constructed according to the parameters, and multi-physics field coupling is realized to simulate the real-time track-vehicle relationship. The model effect is enhanced from five dimensions: physical entity, virtual entity, twin data, connection rules, and application services, thereby obtaining a more realistic five-dimensional model. Step 3: retrieve various parameters of the vehicle when it is about to pass through the section. The prediction of such parameters is obtained through reciprocating actual tests. The parameters are brought into the instability evaluation algorithm to calculate the instability risk of the vehicle on this section. When the result is greater than the preset threshold, it is considered that the train has the risk of instability, and the staff is automatically prompted to take corresponding treatment measures; Step 4: Based on real-time monitoring of various vehicle parameters, data analysis and model reasoning, diagnose the cause of system failure, provide fault location and solutions, and evaluate the system health status based on historical data and model predictions, predict potential failures, and formulate maintenance plans.

Images