Railway vehicle dynamics and structure reliability fusion analysis method and system

Abstract

The invention relates to the technical field of rails, and provides a rail vehicle dynamics and structure reliability fusion analysis method and system, and the method comprises the steps: 1, obtaining the rail operation topological data of a target vehicle, and recognizing a typical operation scene based on line engineering features; 2, constructing an orbit kinetic equation in each typical operation scene; 3, constructing a vehicle body rigid-flexible kinetic equation; 4, coupling the vehicle body rigid-flexible kinetic equation and the track kinetic equation to generate a coupling equation; step 5, obtaining a load curve when the target vehicle passes through the typical track scene, extracting vibration characteristics from the load curve, the vibration characteristics being an equivalent load spectrum; and step 6, inputting the equivalent load spectrum into a pre-established structural reliability fusion analysis model of dynamics performance spatio-temporal evolution, and generating the reliability probability of long-time operation of the vehicle in the line.

Description

Technical Field

[0001] The present application relates to the field of rail technology, and in particular to an analysis method and system for integrating rail vehicle dynamics and structural reliability. Background Art

[0002] The contents of this section merely provide background information related to the present application and may not constitute prior art.

[0003] Rail transit systems are the core components of modern integrated transportation systems, and the service safety of their key equipment directly affects the safe operation of public transportation. In the technological evolution of contemporary rail trains that continuously improve their operating speed and load capacity, core load-bearing components such as car bodies, bogies, wheelsets and couplers continue to bear high-frequency cyclic loads and complex excitations, resulting in progressive degradation of structural dynamic performance. It is worth noting that there is a significant coupling correlation between the cumulative damage effect of components and the instability of dynamic behavior, which may cause major traffic safety accidents with cascading characteristics.

[0004] In current engineering practice, structural reliability assessment and dynamic performance verification of rail vehicles are usually implemented independently as two types of key performance assessment systems: the former verifies the design margin based on static strength criteria, and the latter confirms operational stability through dynamic response analysis. This disciplinary separation leads to dual limitations in traditional reliability analysis: first, static parameters of the initial healthy state are used to characterize the full life cycle performance, and the constitutive relationship between the time-varying load spectrum and material performance degradation is not established; second, the spatial distribution characteristics of dynamic excitation parameters and the cross-scale evolution law of structural fatigue damage are ignored, resulting in the reliability model being established on the basis of quasi-static assumptions. This decoupled analysis method makes it difficult to accurately characterize the nonlinear damage accumulation process caused by the dynamic-static coupling effect during service, resulting in significant theoretical deviations in the safety assessment results. Summary of the invention

[0005] In view of this, the purpose of the present application is to provide an analysis method and system for the fusion of rail vehicle dynamics and structural reliability. The analysis method and system for the fusion of rail vehicle dynamics and structural reliability disclosed in the present application can be realized.

[0006] The purpose of this application is achieved through the following technical solutions: An analysis method for integrating rail vehicle dynamics and structural reliability includes the following steps: Step 1: Obtain the track operation topology data of the target vehicle and identify typical operation scenarios based on line engineering characteristics; Step 2: Construct the orbital dynamics equations for each typical operation scenario; ; Mr represents the track mass matrix, C r represents the track damping matrix, K r represents the track stiffness matrix, x r (t) represents the track displacement vector, F r (t) represents the track external force vector; Step 3: Construct the rigid-flexible dynamics equation of the car body; M ta q + C ta q + K ta q = F ta q; where, M ta represents the vehicle mass matrix, C ta represents the vehicle damping matrix, K ta represents the vehicle stiffness matrix, F ta q represents the vehicle external force vector, q represents the vehicle rigid body degree of freedom vector; Step 4: Couple the rigid-flexible dynamics equation of the car body and the track dynamics equation to generate a coupled equation; The coupled equation is: ; M ta q + C ta q + K ta q = F ta q; Step 5: Obtain the load curve of the target vehicle when passing through a typical track scenario, and extract the vibration characteristics from the load curve, where the vibration characteristics are the equivalent load spectrum; Step 6: Input the equivalent load spectrum into the pre-established structural reliability fusion analysis model of the spatio-temporal evolution of dynamic performance to generate the reliability probability of the vehicle running for a long time on this line.

[0007] In some possible embodiments, Step 2 includes the following steps: Step 21: Divide the track into multiple finite elements, define two nodes at the head and tail of each finite element, and generate a node sequence N1, N2,... N n ; Record the geometric coordinates of each node, and n represents the total number of nodes; Step 22: Collect the track density and track cross-sectional area of the track to construct the track mass matrix M r ; Step 23: Collect the elastic modulus E, Poisson's ratio v, and the vertical support stiffness matrix K of the track bed of the track b , and construct the track stiffness matrix Kr; Step 24: Collect the damping coefficient and modal damping ratio of the track, and calculate the track damping matrix C r ; Step 25: Based on the track mass matrix M r , the track stiffness matrix K r , and the track damping matrix C r to construct the track dynamic equation.

[0008] In some possible embodiments, Step 3 includes the following steps: Step 31: Divide the vehicle into several finite elements, take the head and tail ends of the finite elements as mass points, and generate a sequence of mass points P1, P2, … P PG ; PG represents the total number of mass points; Step 32: Collect the rigid body dynamics parameters, flexible car body parameters, modal reduction parameters, coupling term calculation parameters, and external excitation parameters of the vehicle; Step 33: Construct the vehicle mass matrix M ta ; Step 34: Construct the vehicle stiffness matrix K ta ; Step 35: Construct the vehicle damping matrix C ta ; Step 36: Based on the vehicle mass matrix M ta , the vehicle stiffness matrix K ta , and the vehicle damping matrix C ta to construct the rigid-flexible dynamics equation of the car body.

[0009] In some possible embodiments, Step 21 includes the following steps: Step 211: Collect the track surface information and determine the minimum node density ρ under typical operating scenarios q ; Step 212: Divide the track into N E finite elements, and the length of the finite element is l， Define the head and tail ends of each finite element as nodes; Step 213: Obtain the minimum vibration wavelength λ of the track, and set the finite element length based on the minimum vibration wavelength λ l , where l＜ 10 / λ; Step 214: Continuously adjust the finite element length until the node density ρ0 of the track is equal to the minimum node density ρ q ; Step 215: Traverse the entire track, add new nodes to the worn parts and joint parts of the track, and obtain the node sequence N1, N2, … N n , n represents the total number of nodes; Step 216: Add translational degrees of freedom and rotational degrees of freedom in the x, y, and z directions to each node, and record the geometric coordinates of each node.

[0010] In some possible embodiments, the orbital mass matrix M r is: ; where, I x1 represents the equivalent moment of inertia of the first node about the x-axis, I y1 represents the equivalent moment of inertia of the first node about the y-axis, I z1 represents the equivalent moment of inertia of the first node about the z-axis, I xn represents the equivalent moment of inertia of the nth node about the x-axis, I yn represents the equivalent moment of inertia of the nth node about the y-axis, I zn represents the equivalent moment of inertia of the nth node about the z-axis, m x1 represents the translational mass of the first node on the x-axis, m y1 represents the translational mass of the first node on the y-axis, m z1 represents the translational mass of the first node on the z-axis, m xn represents the translational mass of the nth node on the x-axis, m yn represents the translational mass of the nth node on the y-axis, m zn represents the translational mass of the nth node on the z-axis; diag represents a diagonal matrix; ; ; where, i represents the index of the node, A represents the cross-sectional area of the orbit, ρ represents the orbit density, l e represents the finite element length, e represents the index of the finite element, J represents the torsional constant of the orbit section, I y and I z respectively represent the inertia matrices of the orbit cross-section about the y-axis and z-axis, m i represents the equivalent translational mass of the ith node, I xi represents the equivalent moment of inertia of the ith node about the x-axis, I yi represents the equivalent moment of inertia of the ith node about the y-axis, I zi represents the equivalent moment of inertia of the ith node about the z-axis; The orbital stiffness matrix K r is: ; ; ; where, l e represents the finite element length, e represents the index of the finite element, J represents the torsional constant of the orbit interface, G represents an intermediate variable, ⊕ represents assembling the element matrix into the global matrix according to the degree of freedom numbering, Kb denotes the vertical support stiffness matrix of the ballast bed, E denotes the elastic modulus of the track, v denotes the Poisson's ratio, and I y and I z denote the inertia matrices of the track cross-section about the y-axis and z-axis respectively, and N E denotes the number of finite elements; The track damping matrix is C r : ; ; ; where M r is the track mass matrix, K r is the track stiffness matrix, α1 and β1 are the first track damping coefficient and the second track damping coefficient respectively, are the first damping factor and the second damping factor respectively, is the modal damping ratio.

[0011] In some possible embodiments, the rigid body dynamics parameters include the total vehicle mass, the vehicle material density, the moment of inertia, and the suspension point coordinates; The flexible vehicle parameters include the vehicle material density, the elastic modulus, the shear modulus, the cross-section set parameters, and the mass point division information; The modal reduction parameters include the number of key degrees of freedom, the retained modal order, the suspension stiffness, and the damping; The coupling term calculation parameters include the rigid body Jacobian matrix and the modal shape function. The coupling term calculation parameters select the key degrees of freedom and the retained modal order according to the simulation objective, and are obtained by combining the suspension system parameters.

[0012] The external excitation parameters include the wheel-rail force and the track irregularity excitation.

[0013] In some possible embodiments, the vehicle mass matrix M ta is: ; where M rr denotes the rigid body mass matrix, M rm denotes the inertia coupling matrix, M mm denotes the modal matrix; M mr denotes the transpose matrix of the inertia coupling matrix; ; T denotes the transpose symbol; ; ; ; ; where cI x1Represents the equivalent moment of inertia of the first particle about the x-axis, cI y1 Represents the equivalent moment of inertia of the first particle about the y-axis, cI z1 Represents the equivalent moment of inertia of the first particle about the z-axis, cI xn Represents the equivalent moment of inertia of the nth particle about the x-axis, cI yn Represents the equivalent moment of inertia of the nth particle about the y-axis, cI zn Represents the equivalent moment of inertia of the nth particle about the z-axis, cm x1 Represents the translational mass of the first particle on the x-axis, cm y1 Represents the translational mass of the first particle on the y-axis, cm z1 Represents the translational mass of the first particle on the z-axis, cm xn Represents the translational mass of the nth particle on the x-axis, cm yn Represents the translational mass of the nth particle on the y-axis, cm zn Represents the translational mass of the nth particle on the z-axis; diag represents a diagonal matrix; Represents the transformation matrix that maps physical coordinates to modal coordinates, M represents the un-reduced global mass matrix of the flexible car body, Represents the volume integral of the entire car body structure, R is the free quantity matrix, 0 is the zero matrix of PB×L, L is the number of retained elastic modal orders, Is the boundary-induced displacement transfer matrix, PB is the number of key degrees of freedom; ; K II Represents the internal degrees of freedom stiffness matrix, K IB Represents the internal-boundary degrees of freedom coupling stiffness matrix, Represents the fixed interface modal matrix; ; M II Represents the internal degrees of freedom mass matrix, Represents the diagonal matrix of the first K eigenvalues; ρ represents the density of the car body material, Represents the modal shape function matrix, JD represents the Jacobian matrix of rigid body motion; The vehicle stiffness matrix is K ta : ; K rr Represents the vehicle rigid body stiffness matrix, K mm Represents the modal stiffness matrix, K rm Represents the stiffness coupling matrix, K mr Is the transpose matrix of K rm ; ; ; ; T r represents the transformation matrix from the rigid body displacement to the suspension point, K s represents the local stiffness matrix of the suspension system, represents the transformation matrix that maps physical coordinates to modal coordinates, K represents the un-reduced overall stiffness matrix of the vehicle, T p represents the transformation matrix from modal coordinates to the suspension point; Vehicle damping matrix C ta is: ; wherein, C rr represents the damping matrix, C rm represents the damping coupling matrix, C mm represents the modal damping matrix, C mr represents the transpose matrix of the damping coupling matrix; wherein, ; ; ;

[0014] T r represents the transformation matrix from the rigid body displacement to the suspension point, C s represents the suspension damping, α2 and β2 are the first vehicle damping coefficient and the second vehicle damping coefficient respectively, T p represents the transformation matrix from modal coordinates to the suspension point, T represents the transpose matrix.

[0015] In some possible embodiments, the typical operating scenarios include the operating route with the maximum turning radius, the uphill route with the maximum gradient, the downhill route with the maximum gradient, and the operating route at the highest speed.

[0016] In some possible embodiments, step 5 includes the following steps: Step 51: Input the coupled equation into the DAP-Rail software; Step 52: Input the operating scenario parameters, and the operating scenario parameters include line parameters, vehicle parameters, and load conditions; Step 53: The DAP-Rail software outputs a load curve corresponding to time, and extracts an equivalent load spectrum from the load curve.

[0017] In some possible embodiments, step 53 includes the following steps: Step 531: Filter the load curve with a high-pass filter to obtain the time series X(t); Step 532: Identify the maximum and minimum values of the time series X(t), and extract the local maximum values and local minimum values ; represents the s-th local maximum value, Denote the s-th local minimum value and generate an extreme value sequence ST; ST = ; Step 533: Starting from the first value in the extreme value sequence ST, gradually traverse the entire extreme value sequence ST. When the preset closing condition is met, extract a cycle SY, and record the amplitude SY1 and mean value SY2 of this cycle; ; ; Extract all cycles SY and generate a cycle list; Step 534: Define the amplitude binning interval, and allocate the amplitude of each cycle SY to the corresponding amplitude binning interval; output the binning count table; Step 535: Obtain the amplitude SY1, number of cycles Q, and material S-N curve parameters of each cycle, generate an equivalent constant amplitude load amplitude, and convert the equivalent constant amplitude load amplitude into an equivalent load spectrum; ; where BY represents the number of amplitude binning intervals, b represents the index of the amplitude binning interval, B b represents the number of cycles in the b-th amplitude binning interval, SY1 b represents the maximum amplitude in the b-th amplitude binning interval, mba represents the slope of the material S-N curve; N total represents the sum of the number of cycles.

[0018] In some possible embodiments, Step 6 includes the following steps: Step 61: Construct an LSTM model to extract time series features from the equivalent load spectrum; Step 62: Construct two sub-network groups with the same structure and shared weights; Step 63: Pre-construct a database related to time series features and reliability probability; Step 64: Input the time series features to be predicted into one sub-network group, and gradually input the time series features in the database into the other sub-network group, calculate the similarity between the time series features and the time series features in the database, screen out the sample with the highest similarity from the database, and obtain the reliability probability of this time series feature.

[0019] An analysis system for the integration of railway vehicle dynamics and structural reliability generates the reliability probability of the vehicle running on this line for a long time by using the foregoing analysis method for the integration of railway vehicle dynamics and structural reliability.

[0020] The technical solution of the embodiments of the present application has at least the following advantages and beneficial effects: (1)In the technical solution proposed in this application, aiming at the problem of reliability probability assessment of the railway vehicle system, a multi-physical field coupling dynamics model of wheel-rail contact is established, and the dynamic load history of the vehicle under all working conditions is obtained through numerical solution. Based on the nonlinear random vibration theory, the multi-axial equivalent fatigue load spectrum is derived, and combined with the material S-N curve and the Miner linear cumulative damage criterion, a time-varying reliability assessment function is constructed. This method breaks through the limitations of traditional static reliability analysis. By analyzing the frequency-domain energy distribution characteristics of the load spectrum, it effectively identifies the cross-domain correlation of stress amplitude - cycle number - phase difference in the structural dynamic response, and realizes the accurate prediction of the initiation and propagation rate of micro-cracks within the entire life cycle.

[0021] (2)When constructing the wheel-rail coupling dynamics model in this solution, the rigid-flexible hybrid multi-body system modeling method is adopted, and the two-way coupling of rigid body motion and elastic deformation is realized by introducing the flexible body modal neutral file. Compared with the traditional pure rigid body model or the decoupled rigid-flexible separation model, this method completely retains the geometric nonlinear terms and inertial coupling terms in the dynamic equation, can accurately characterize the modulation effect of the structural dynamic flexibility on the wheel-rail contact force under high-frequency excitation, and significantly improves the time-frequency domain prediction accuracy of the wheel-rail impact load.

[0022] (3)In the rigid-flexible coupling modeling of the car body, the degrees of freedom are compressed. By retaining the rigid body degrees of freedom of the interface main nodes and the low-order elastic modes, the original distributed parameter model with 6N + 6 degrees of freedom is reduced to a condensed model with K + 6 degrees of freedom. While ensuring the accuracy of the first six-order rigid body motion, this reduction algorithm effectively suppresses the truncation error of the high-order modes, and maintains a kinetic energy conservation rate of more than 99.5% while reducing the calculation scale by two orders of magnitude.

[0023] (4)In terms of dynamic excitation modeling, a modal space projection algorithm for distributed loads is proposed. By constructing a load influence coefficient matrix, complex excitations such as wheel-rail contact forces and aerodynamic loads are accurately mapped to the generalized modal coordinates. This algorithm breaks through the limitations of traditional concentrated force loading and can accurately characterize the wave propagation effect and strain gradient distribution characteristics on the surface of the elastic body under high-frequency vibration.

[0024] (5)During the finite element mesh discretization process, a mesh convergence criterion based on wave theory is established, and the critical mesh size is determined according to the longitudinal wave / transverse wave propagation speed of the material and the highest frequency component of the excitation. By constructing a reference division scale, while suppressing the numerical dispersion effect, the mesh density is reduced by 40% - 60% compared with the traditional empirical method, and the accurate analysis of the micro-strain gradient is realized.

[0025] (6) In this solution, when generating the equivalent load spectrum, the time history of the random load is converted into an equivalent constant amplitude or multi-level load block to ensure the same fatigue damage as the original load, avoiding the defect of ignoring the non-linear damage contribution of high-amplitude loads in traditional methods. According to the S-N curve parameters, the damage weights of different load amplitudes are quantified, which can well integrate the damage characteristics of the material and dynamically display the load distribution and development of the material under long-term operation, thus accurately generating the equivalent load diagram.

[0026] (7) This application uses a high-pass filter to filter out low-frequency and invalid information in the load curve, reducing the influence of error information and noise information on the equivalent load diagram.

[0027] (8) This application uses a siamese network model for similarity matching. Compared with the scheme using cosine similarity matching, in this application, the similarity analysis situation can be generated from the perspective of the relationship between features. Compared with the pure numerical comparison, the similarity in this scheme can better consider the influence between features and the evolution direction of features, so that higher recognition accuracy can be achieved with a smaller data sample.

[0028] (9) In the technical solution provided by this application, the track constructs the orbital dynamics equation, and the vehicle constructs the rigid-flexible dynamics equation. The reason is that when the vehicle is running, there will be a resonance phenomenon between the flexible part of the vehicle and the flexible part of the track. Introducing flexible information into both will increase the computational amount of the model and the simulation difficulty, but reduce the accuracy. In this application, only the flexible characteristics of the vehicle are considered. The reason is that the vehicle integrates a suspension system and has more connection points, so the flexible influence of the vehicle is greater. By establishing the rigid-flexible dynamics equation of the vehicle and then coupling the rigid-flexible dynamics equation of the vehicle with the orbital dynamics equation, the flexible part in the vehicle-rail system can be well described, increasing the simulation ability of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic structural diagram of an analysis method for integrating orbital vehicle dynamics and structural reliability provided by some embodiments of this application.

[0030] Figure 2 It is a rendering of the track in the DAP-Rail software.

[0031] Figure 3 It is a rendering of the vehicle in the DAP-Rail software.

[0032] Figure 4 It is a rendering of the track and the vehicle in DAP-Rail.

[0033] Figure 5It is a load curve graph, where the abscissa is time in seconds and the ordinate is the load ratio.

[0034] Figure 6 It is an equivalent load spectrum. Specific implementation manners

[0035] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific implementation manners. The same reference numerals in the drawings represent the same components. It should be noted that the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0036] Compared with the embodiments shown in the drawings, the feasible implementation solutions within the scope of protection of the present application may have fewer components, have other components not shown in the drawings, different components, differently arranged components or differently connected components, etc. In addition, two or more components in the drawings may be implemented in a single component, or a single component shown in the drawings may be implemented as multiple separate components.

[0037] Unless otherwise defined, the technical terms or scientific terms used herein shall have the ordinary meanings understood by those of ordinary skill in the art to which the present application pertains. The "first", "second" and similar terms used in the specification and claims of the present application do not denote any order, quantity or importance, but are only used to distinguish different components. Similarly, terms such as "a" or "one" do not necessarily denote a quantity limitation. "Up", "down", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly. Embodiment 1:

[0038] Reference Figure 1 The first embodiment of the present application discloses an analysis method for integrating the dynamics and structural reliability of rail vehicles, including the following steps: Step 1: Obtain the track operation topology data of the target vehicle, and identify typical operation scenarios based on the line engineering characteristics. The typical operation scenarios include the operation route with the maximum turning radius, the uphill route with the maximum gradient, the downhill route with the maximum gradient, and the operation route at the highest speed. The line engineering characteristics are the track turning radius, the track speed limit, and the track gradient.

[0039] In the field of track engineering, the full-line simulation of locomotives and rolling stock faces the problem of the computational complexity of multi-physical field coupling: the dynamic structural stability analysis of a complete line requires the construction of a global time-varying load model, resulting in exponential consumption of computing resources and the risk of iterative error propagation. The present invention proposes a dimensionality reduction analysis method based on the characteristics of track engineering, and constructs a typical operation scenario by extracting the key sections with the most structural dynamics bearing characteristics in the line. It is demonstrated from the perspective of engineering mechanics that when the vehicle meets the structural reliability threshold under boundary conditions such as maximum curvature, maximum slope direction, and limit speed, the dynamic response envelope must cover the threshold of the conventional working condition, thereby establishing a mathematical completeness proof of the structural integrity of the entire line. This screening mechanism based on characteristic working conditions not only maintains the engineering accuracy of the dynamic simulation but also realizes a step-by-step improvement in computing efficiency.

[0040] Step 2: Construct the track dynamics equation for each typical operation scenario; ; M r represents the track mass matrix, C r represents the track damping matrix, K r represents the track stiffness matrix, x r (t) represents the track displacement vector, F r (t) represents the track external force vector.

[0041] For each typical operation scenario, the structure of its track is different, and there will be certain differences when constructing the track dynamics equation accordingly. Therefore, the track dynamics equation for each typical operation scenario needs to be established independently.

[0042] Furthermore, Step 2 includes the following steps: Step 21: Divide the track into multiple finite elements, define two nodes at the head and tail of each finite element, and generate a node sequence N1, N2,...; record the geometric coordinates of each node.

[0043] Specifically, Step 21 includes the following steps: Step 211: Collect the track surface information and determine the minimum node density ρ q .

[0044] When constructing the track dynamics equation, the discretized node scale is positively correlated with the model solution accuracy but will lead to a significant increase in computational complexity. In view of this, this solution innovatively introduces an adaptive node density configuration strategy, establishes a gradient grid division criterion for different characteristic working conditions, and realizes the dynamic balance between computational accuracy and efficiency.

[0045] At the engineering implementation level, a node density regulation function is established based on the health status of the track surface structure: taking the historical maintenance frequency of track units as the key input parameter, there is a monotonically increasing relationship between the maintenance frequency threshold and the node distribution density. Engineering practice shows that the track subgrade will exhibit differential wear distribution characteristics under the action of the wheel-rail contact stress field. The high-frequency maintenance section often corresponds to the area of accumulated plastic deformation of materials or the area where contact fatigue cracks initiate. In numerical simulation, it is necessary to improve the discretization accuracy to accurately characterize its mechanical response characteristics. By establishing a quantitative mapping model among maintenance records - structural degradation sensitivity - grid resolution, ensure that the simulation accuracy of key weak areas meets the requirements of micron-level deformation resolution.

[0046] Step 212: Divide the track into N E finite elements, and the length of the finite element is l， Define the head and tail ends of each finite element as nodes; Step 213: Obtain the minimum vibration wavelength λ of the track, and set the finite element length based on the minimum vibration wavelength λ l where l＜ 10 / λ; Step 214: Continuously adjust the finite element length until the node density ρ0 of the track is equal to the minimum node density ρ q .

[0047] In the technical solution proposed in this application, during the mesh discretization process, the finite element unit division scale is determined based on the minimum characteristic wavelength of the excitation wave spectrum, and it is ensured that the mesh density meets the critical convergence threshold. This optimization method strictly follows the convergence criterion of wave theory. On the premise of ensuring the modal truncation error, the optimal compression of the calculation scale is achieved. This dual constraint mechanism not only effectively suppresses the Gibbs phenomenon of high-frequency vibration signals but also enables the numerical model to accurately capture the micron-level strain gradient distribution characteristics through adaptive adjustment of the unit topological structure after reducing the order dimension, achieving the coordinated improvement of computational resource optimization and dynamic fidelity.

[0048] Step 215: Traverse the entire track, add new nodes to the worn part and joint part of the track, and obtain the node sequence N1, N2,... N n , where n represents the total number of nodes.

[0049] For relatively special positions, more nodes need to be added to supplement the details of the model. For the identification of the worn area of the track, an edge inspection algorithm can be used to automatically add new nodes.

[0050] Step 216: Add translational degrees of freedom and rotational degrees of freedom in the x, y, and z directions to each node, and record the geometric coordinates of each node.

[0051] In this way, each node has 6 degrees of freedom.

[0052] Step 22: Collect the orbital density and orbital cross-sectional area of the orbit to construct the orbital mass matrix M r ; The essence of the mass matrix is to evenly distribute the mass in the orbit to each node. Specifically, the orbital mass matrix M r is: ; where, I x1 represents the equivalent moment of inertia of the first node about the x-axis, I y1 represents the equivalent moment of inertia of the first node about the y-axis, I z1 represents the equivalent moment of inertia of the first node about the z-axis, I xn represents the equivalent moment of inertia of the nth node about the x-axis, I yn represents the equivalent moment of inertia of the nth node about the y-axis, I zn represents the equivalent moment of inertia of the nth node about the z-axis, m x1 represents the translational mass of the first node on the x-axis, m y1 represents the translational mass of the first node on the y-axis, m z1 represents the translational mass of the first node on the z-axis, m xn represents the translational mass of the nth node on the x-axis, m yn represents the translational mass of the nth node on the y-axis, m zn represents the translational mass of the nth node on the z-axis; diag represents a diagonal matrix; ; ; where, i represents the index of the node, A represents the orbital cross-sectional area, ρ represents the orbital density, l e represents the finite element length, e represents the index of the finite element, J represents the orbital section torsion constant, I y and I z respectively represent the inertia matrices of the orbital cross-section about the y-axis and z-axis, m i represents the equivalent translational mass of the ith node, I xi represents the equivalent moment of inertia of the ith node about the x-axis, I yi represents the equivalent moment of inertia of the ith node about the y-axis, I zi represents the equivalent moment of inertia of the ith node about the z-axis, where, m i =m xi =m yi =m zi ; Among them, the track interface torsional constant, and the moments of inertia of the track cross-section about the y-axis and z-axis are pre-calculated from the surface data, which are data that need to be collected in advance. After collecting the above information, the mass and inertia characteristics of the track can be incorporated into the corresponding nodes.

[0053] Step 23: Collect the elastic modulus E, Poisson's ratio v, and the vertical support stiffness matrix K of the track bed b , and construct the track stiffness matrix K r ; ; ; ; Among them, l e represents the finite element length, e represents the finite element index, J represents the track interface torsional constant, G represents an intermediate variable, ⊕ represents assembling the element matrix into the global matrix according to the degree of freedom numbering, I y and I z respectively represent the inertia matrices of the track cross-section about the y-axis and z-axis, N E represents the number of finite elements.

[0054] Step 24: Collect the damping coefficient and modal damping ratio of the track, and calculate the track damping matrix C r ; ; ; ; Among them, M r is the track mass matrix, K r is the track stiffness matrix, α1 and β1 are the first track damping coefficient and the second track damping coefficient respectively, are the first damping factor and the second damping factor respectively; is the modal damping ratio.

[0055] Step 25: Construct the track dynamic equation according to the track mass matrix M r , the track stiffness matrix K r , and the track damping matrix C r .

[0056] In this way, in Step 2, the above steps can be used to construct a corresponding track dynamic equation for each typical operating scenario.

[0057] Step 3: Construct the rigid-flexible dynamic equation of the car body; M ta q + C ta q + K ta q = F ta q; Among them, M ta represents the vehicle mass matrix, C ta represents the vehicle damping matrix, K ta represents the vehicle stiffness matrix, F ta q represents the vehicle external force vector, and q represents the vehicle rigid body degree of freedom vector.

[0058] Furthermore, step 3 includes the following steps: Step 31: Divide the vehicle into several finite elements, take the head and tail ends of the finite elements as mass points, and generate a sequence of mass points P1, P2, … P PG ; PG represents the total number of mass points; In step 31, it is mainly necessary to divide the vehicle into several finite elements, then return the mass within the finite elements to the corresponding mass points, and simplify the vehicle into a model composed of mass points. The method of dividing the vehicle into mass points is the same as the method of dividing the track into nodes, which will not be elaborated here.

[0059] Step 32: Collect the rigid body dynamics parameters, flexible car body parameters, modal reduction parameters, coupling term calculation parameters, and external excitation parameters of the vehicle; Among them, the rigid body dynamics parameters include the total vehicle mass, vehicle body material density, moment of inertia, and suspension point coordinates; the rigid body dynamics parameters are extracted from the CAD model or test report.

[0060] The flexible car body parameters include the vehicle body material density, elastic modulus, shear modulus, cross-sectional set parameters, and mass point division information; the flexible car body parameters are obtained through finite element preprocessing software.

[0061] The modal reduction parameters include the number of key degrees of freedom, the number of retained modal orders, suspension stiffness, and damping.

[0062] The coupling term calculation parameters include the rigid body Jacobian matrix and modal shape function. The coupling term calculation parameters select the key degrees of freedom and the number of retained modal orders according to the simulation objective, and obtain them in combination with the suspension system parameters.

[0063] The external excitation parameters include wheel-rail force and track irregularity excitation. The external excitation parameters are loaded and obtained from the multi-body dynamics model or measured data.

[0064] Using the rigid body dynamics parameters, flexible car body parameters, modal reduction parameters, coupling term calculation parameters, and external excitation parameters, the elastic modal order, internal-boundary degree of freedom coupling stiffness matrix, fixed interface modal matrix, internal degree of freedom mass matrix, eigenvalue diagonal matrix, modal shape function matrix, rigid body displacement to suspension point transformation matrix, local stiffness matrix of the suspension system, transformation matrix from physical coordinates to modal coordinates, vehicle overall stiffness matrix, and modal coordinate to suspension point transformation matrix can be generated.

[0065] Step 33: Construct the vehicle mass matrix M ta ; ; where M rr represents the rigid body mass matrix, M rm represents the inertia coupling matrix, M mm represents the modal matrix; M mr represents the transpose matrix of the inertia coupling matrix; ; T represents the transpose symbol; ; ; ; ; where cI x1 represents the equivalent moment of inertia of the first particle about the x-axis, cI y1 represents the equivalent moment of inertia of the first particle about the y-axis, cI z1 represents the equivalent moment of inertia of the first particle about the z-axis, cI xn represents the equivalent moment of inertia of the nth particle about the x-axis, cI yn represents the equivalent moment of inertia of the nth particle about the y-axis, cI zn represents the equivalent moment of inertia of the nth particle about the z-axis, cm x1 represents the translational mass of the first particle in the x-axis, cm y1 represents the translational mass of the first particle in the y-axis, cm z1 represents the translational mass of the first particle in the z-axis, cm xn represents the translational mass of the nth particle in the x-axis, cm yn represents the translational mass of the nth particle in the y-axis, cm zn represents the translational mass of the nth particle in the z-axis; diag represents the diagonal matrix; represents the transformation matrix that maps physical coordinates to modal coordinates, M represents the un-reduced overall mass matrix of the flexible vehicle body, represents the volume integral of the entire vehicle body structure, dV represents the integral symbol, R is the free quantity matrix, 0 is the zero matrix of PB×L, L is the number of retained elastic modal orders, is the boundary-induced displacement transfer matrix, PB is the number of key degrees of freedom; ; K II represents the internal degrees of freedom stiffness matrix, K IB represents the internal-boundary degrees of freedom coupling stiffness matrix, represents the fixed interface modal matrix; ; M II represents the internal degree-of-freedom mass matrix, represents the diagonal matrix of the first K eigenvalues; ρ represents the density of the vehicle body material, represents the modal shape function matrix, and JD represents the Jacobian matrix of rigid body motion.

[0066] Step 34: Construct the vehicle stiffness matrix K ta ; ; K rr represents the vehicle rigid body stiffness matrix, K mm represents the modal stiffness matrix, K rm represents the stiffness coupling matrix, K mr is the transpose matrix of K rm ; ; ; ; T r represents the transformation matrix from rigid body displacement to the suspension point, K s represents the local stiffness matrix of the suspension system, represents the transformation matrix that maps physical coordinates to modal coordinates, K represents the un-reduced overall vehicle stiffness matrix, T p represents the transformation matrix from modal coordinates to the suspension point.

[0067] Assume that the vehicle body has 10 mass points, the total degrees of freedom are 66, the key degrees of freedom are 8, and 10 modes are retained: , which describes the relationship between the displacements of 8 suspension points and 6 rigid body degrees of freedom.

[0068] , the reduced modal stiffness matrix.

[0069] , the stiffness coupling that connects rigid body motion and 10 elastic modes.

[0070] Through the above parameter definitions, the rigid-flexible coupling dynamic equation can be systematically constructed to realize the co-simulation of vehicle vibration and rigid body motion.

[0071] Step 35: Construct the vehicle damping matrix C ta ; ; where C rr represents the damping matrix, C rm represents the damping coupling matrix, C mm represents the modal damping matrix, C mr represents the transpose matrix of the damping coupling matrix; Among them, ; ; ; T r represents the transformation matrix from the rigid body displacement to the suspension point, C s represents the suspension damping, α2 and β2 are the first vehicle damping coefficient and the second vehicle damping coefficient respectively, T p represents the transformation matrix from the modal coordinates to the suspension point, and T represents the transpose matrix.

[0072] Step 36: Construct the rigid-flexible dynamics equation of the vehicle body according to the vehicle mass matrix M ta , the vehicle stiffness matrix K ta , the vehicle damping matrix C ta ; ; Among them, q represents the rigid body degree of freedom vector of the vehicle, including translational and rotational components; F r represents the external force vector of the rigid body degree of freedom, F m represents the external force vector of the modal coordinates; ; represents the modal shape function matrix, f ext represents the external distributed force density, represents the volume integral of the entire vehicle body structure, and dV represents the integral symbol.

[0073] Step 4: Couple the rigid-flexible dynamics equation of the vehicle body and the track dynamics equation to generate a coupled equation; The coupled equation is: ; M ta q + C ta q + K ta q = F ta q; Among them, M r represents the track mass matrix, C r represents the track damping matrix, K r represents the track stiffness matrix, x r (t) represents the track displacement vector, F r (t) represents the track external force vector; M ta represents the vehicle mass matrix, C ta represents the vehicle damping matrix, K ta represents the vehicle stiffness matrix, q represents the rigid body degree of freedom vector of the vehicle, F ta q represents the vehicle external force vector.

[0074] Reference Figures 2 to 4, Step 5: Obtain the load curve of the target vehicle when passing through a typical track scenario, and extract the vibration characteristics from the load curve. The vibration characteristics are the equivalent load spectrum.

[0075] The equivalent load spectrum needs to be obtained through a coupling equation. The specific method is as follows: Step 51: Input the coupling equation into the DAP-Rail software.

[0076] Step 52: Input the operation scenario parameters, which include line parameters, vehicle parameters, and load conditions.

[0077] The line parameters include gradient, curve radius, and track spectrum; the vehicle parameters include speed and axle load; the load conditions include tractive force and braking force.

[0078] Reference Figure 5 , Step 53: The DAP-Rail software outputs the load curve corresponding to time, and extracts the equivalent load spectrum from the load curve. Figure 5 In [reference], the abscissa is time, with the unit of second, and the ordinate is the load ratio KU. KU = (Nk - N0)Nk / N0, where N0 represents the standard load, N0 = 180 kN, and Nk represents the load at the current time point. When KU is greater than zero, it means the current load is pressure, and when KU is less than zero, it means the current load is tension.

[0079] Reference Figure 6 , Further, Step 53 includes the following steps: Step 531: Filter the load curve with a high-pass filter to obtain the time series X(t); There is a lot of noise information in the load curve. Therefore, it needs to be filtered with a high-pass filter. The high-pass filter will filter out the information with lower frequencies. In this way, the information with higher frequencies can be retained. In practice, the probability of these information with higher frequencies belonging to the correct information is higher, and it is more valuable to retain them.

[0080] Step 532: Identify the maximum and minimum values of the time series X(t), and extract the local maximum values and local minimum values ; represents the s-th local maximum value, represents the s-th local minimum value, and generate the extreme value sequence ST; ST = ; Step 533: Starting from the first value in the extreme value sequence ST, gradually traverse the entire extreme value sequence ST. When the preset closing condition is met, extract a cycle SY, and record the amplitude SY1 and mean value SY2 of this cycle; ; ; Extract all cyclic SYs to generate a cyclic list; Step 534: Define the amplitude binning intervals, and assign the amplitude of each cyclic SY to the corresponding amplitude binning interval; output the binning count table; Step 535: Obtain the amplitude SY1, number of cycles Q, and material S-N curve parameters of each cycle, generate an equivalent constant amplitude load amplitude, and convert the equivalent constant amplitude load amplitude into an equivalent load spectrum; ; where BY represents the number of amplitude binning intervals, b represents the index of the amplitude binning interval, B b represents the number of cycles in the b-th amplitude binning interval, SY1 b represents the maximum amplitude in the b-th amplitude binning interval, mba represents the slope of the material S-N curve; N total represents the sum of the number of cycles.

[0081] The equivalent load spectrum is to convert the complex random load time history into a simplified and standardized load sequence through the fatigue damage equivalence principle. Given the equivalent constant amplitude load amplitude, converting the equivalent load spectrum is prior art and will not be elaborated here.

[0082] Step 6: Input the equivalent load spectrum into a pre-established structural reliability fusion analysis model for the spatio-temporal evolution of dynamic performance to generate the reliability probability of the vehicle during long-term operation on this line.

[0083] Step 6 includes the following steps: Step 61: Construct an LSTM model to extract temporal features from the equivalent load spectrum.

[0084] The LSTM model includes an input layer, an LSTM layer, and an output layer; the input layer is used to input the equivalent load spectrum, the LSTM layer is used to extract the temporal characteristics in the equivalent load spectrum, and the output layer is used to output the temporal features corresponding to the temporal characteristics.

[0085] The LSTM model in this solution essentially converts the information of an equivalent load spectrum into the temporal features of a temporal sequence. Compared with the spectrogram, the information density of the temporal features is higher.

[0086] Step 62: Construct two sub-network groups with the same structure and weight sharing; The sub-network group is a CNN convolutional network model with the same structure, input, and output. Using two sub-network groups with the same structure can better extract the hidden information in the temporal features, identify the actual differences between the temporal features, and thus accurately predict the reliability probability.

[0087] Step 63: Pre - construct a database related to temporal features and reliability probabilities.

[0088] The database stores a large number of samples, and each sample includes a set of temporal features and a reliability probability. The samples are obtained through actual experiments.

[0089] Step 64: Input the temporal features to be predicted into a sub - network group, and gradually input the temporal features in the database into another sub - network group. Calculate the similarity between the temporal features and the temporal features in the database, screen out the sample with the highest similarity from the database, and obtain the reliability probability of this temporal feature.

[0090] The reason why two sub - network groups with the same structure and shared weights are needed in this application to calculate the similarity between samples is that the samples are difficult to obtain and there are large differences between samples. In practice, it is impossible to obtain a sufficient number of samples. Therefore, it is necessary to accurately analyze the internal differences between temporal features. For this reason, using two sub - network groups with the same structure and shared weights to calculate similarity in this solution can better capture the differences between features, and then be able to regress the input temporal features to the corresponding reliability probabilities with a smaller number of samples.

[0091] The structural reliability fusion analysis model for the spatio - temporal evolution of dynamic performance in this application consists of two sub - network groups with the same structure and shared weights, and a corresponding database related to temporal features and reliability probabilities.

[0092] Furthermore, the sub - network group needs to be trained before use. When training the sub - network group, the loss function is: ; where, represents the loss function, NU represents the number of training samples, il represents the index of the training sample, y il represents the il th training sample's true label, P il represents the il th training sample's reliability probability, and log represents the logarithmic function.

[0093] Embodiment 2: An analysis system for the fusion of the dynamics and structural reliability of a rail vehicle uses the aforementioned analysis method for the fusion of the dynamics and structural reliability of a rail vehicle to generate the reliability probability of the vehicle during long - term operation on this line.

[0094] The above are only the preferred embodiments of this application and are not used to limit this application. For those skilled in the art, this application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included in the protection scope of this application.

Claims

1. An analysis method for integrating rail vehicle dynamics and structural reliability, characterized in that: The steps include: Step 1: Obtain the track operation topology data of the target vehicle and identify typical operation scenarios based on line engineering characteristics; Step 2: Construct the orbital dynamics equations for each typical operation scenario; ； Where M r represents the orbital mass matrix, C r represents the track damping matrix, K r represents the track stiffness matrix, x r (t) represents the orbital displacement vector, F r (t) represents the orbital external force vector; Step 3: Construct the rigid-flexible dynamic equation of the vehicle body; M ta q+C ta q+K ta q=F ta q; Among them, M ta represents the vehicle mass matrix, C ta represents the vehicle damping matrix, K ta represents the vehicle stiffness matrix, F ta q represents the vehicle external force vector, q represents the vehicle rigid body freedom vector; Step 4: Couple the vehicle rigid-flexible dynamics equation and the track dynamics equation to generate a coupling equation; The coupling equations are: ； M ta q+C ta q+K ta q=F ta q; Step 5: Obtain a load curve of the target vehicle when passing through a typical track scene, and extract a vibration feature from the load curve, wherein the vibration feature is an equivalent load spectrum; Step 6: Input the equivalent load spectrum into the pre-established structural reliability fusion analysis model of the spatiotemporal evolution of dynamic performance to generate the reliability probability of the vehicle running on the line for a long time.

2. The analysis method for integrating rail vehicle dynamics and structural reliability according to claim 1 is characterized in that: Step 2 includes the following steps: Step 21: Divide the track into multiple finite elements, define two nodes at the beginning and end of each finite element, and generate a node sequence N1, N2, ...N n ;Record the geometric coordinates of each node, n represents the total number of nodes; Step 22: Collect the track density and track cross-sectional area of ​​the track to construct the track mass matrix M r ; Step 23: Collect the track elastic modulus E, Poisson's ratio v, and the vertical support stiffness matrix K of the track b , construct the track stiffness matrix Kr; Step 24: Collect the track damping coefficient and modal damping ratio, and calculate the track damping matrix C r ; Step 25: According to the orbital mass matrix M r , track stiffness matrix K r , track damping matrix C r Construct the orbital dynamics equations.

3. The analysis method for integrating rail vehicle dynamics and structural reliability according to claim 1 is characterized in that: Step 3 includes the following steps: Step 31: Divide the vehicle into several finite elements, take the first and last ends of the finite element as mass points, and generate a mass point sequence P1, P2, ...P PG ;PG represents the total number of particles; Step 32: Collect rigid body dynamic parameters, flexible body parameters, modal reduction parameters, coupling term calculation parameters and external excitation parameters of the vehicle; Step 33: Construct the vehicle mass matrix M ta ; Step 34: Construct the vehicle stiffness matrix K ta ; Step 35: Construct the vehicle damping matrix C ta ; Step 36: According to the vehicle mass matrix M ta , vehicle stiffness matrix K ta , vehicle damping matrix C ta Construct the rigid-flexible dynamic equation of the vehicle body.

4. The analysis method for integrating rail vehicle dynamics and structural reliability according to claim 2 is characterized in that: Step 21 includes the following steps: Step 211: Collect track surface information and determine the minimum node density ρ under typical operation scenarios q ; Step 212: Divide the track into N E finite elements, the length of the finite element is l， Define the first and last ends of each finite element as nodes; Step 213: Obtain the minimum vibration wavelength λ of the track, and set the finite element length based on the minimum vibration wavelength λ l ,in l＜ 10 / λ; Step 214: Continue to adjust the finite element length until the node density ρ0 of the track is equal to the minimum node density ρ q ; Step 215: traverse the entire track, add new nodes to the worn parts and joint parts of the track, and obtain the node sequence N1, N2, ...N of the track n , n represents the total number of nodes; Step 216: For each node, add the translational degrees of freedom and rotational degrees of freedom in the x, y, and z directions, and record the geometric coordinates of each node.

5. The analysis method for integrating rail vehicle dynamics and structural reliability according to claim 2 is characterized in that: Track mass matrix M r for: ； Among them, I x1 represents the equivalent moment of inertia of the first node around the x-axis, I y1 represents the equivalent moment of inertia of the first node around the y-axis, I z1 represents the equivalent moment of inertia of the first node around the z-axis, I xn represents the equivalent moment of inertia of the nth node around the x-axis, I yn represents the equivalent moment of inertia of the nth node around the y-axis, I zn represents the equivalent moment of inertia of the nth node around the z-axis, m x1 Represents the translational mass of the first node on the x-axis, m y1 Represents the translational mass of the first node on the y-axis, m z1 Represents the translational mass of the first node on the z-axis, m xn represents the translational mass of the nth node on the x-axis, m yn represents the translational mass of the nth node on the y-axis, m zn represents the translational mass of the nth node in the z-axis; diag represents the diagonal matrix; ； ； Where i represents the index of the node, A represents the track cross-sectional area, and ρ represents the track density. l e represents the finite element length, e represents the finite element index, J represents the track section torsion constant, I y and I z Respectively represent the inertia matrix of the track cross section around the y-axis and z-axis, m i represents the equivalent translational mass of the i-th node, I xi represents the equivalent moment of inertia of the ith node around the x-axis, I yi represents the equivalent moment of inertia of the ith node around the y-axis, I zi represents the equivalent moment of inertia of the ith node around the z-axis; Track stiffness matrix K r for: ； ； ； in, l e represents the finite element length, e represents the index of the finite element, J represents the orbital interface torsion constant, G represents the intermediate variable, ⊕ represents assembling the unit matrix into the global matrix according to the degree of freedom number, K b represents the vertical support stiffness matrix of the track bed, E represents the elastic modulus of the track, v represents the Poisson's ratio, I y and I z Respectively represent the inertia matrix of the track cross section around the y-axis and z-axis, N E represents the number of finite elements; The track damping matrix is ​​C r : ； ； ； Among them, M r is the orbital mass matrix, K r is the track stiffness matrix, α1 and β1 are the first track damping coefficient and the second track damping coefficient respectively, are the first damping factor and the second damping factor, respectively. is the modal damping ratio.

6. The analysis method for integrating rail vehicle dynamics and structural reliability according to claim 3 is characterized in that: The rigid body dynamics parameters include the total mass of the vehicle body, the density of the vehicle body material, the moment of inertia, and the coordinates of the suspension points; Flexible body parameters include body material density, elastic modulus, shear modulus, section set parameters, and mass point division information; Modal reduction parameters include the number of critical degrees of freedom, the number of retained modal orders, suspension stiffness, and damping; The coupling term calculation parameters include the rigid body Jacobian matrix and the modal shape function. The coupling term calculation parameters select the key degrees of freedom and the reserved modal order according to the simulation objectives. External excitation parameters include wheel-rail force and track unevenness excitation.

7. The analysis method for integrating rail vehicle dynamics and structural reliability according to claim 6 is characterized in that: Vehicle mass matrix M ta for: ； Among them, M rr represents the rigid body mass matrix, M rm represents the inertial coupling matrix, M mm represents the modal matrix; M mr represents the transposed matrix of the inertial coupling matrix; ; T represents the transposition symbol; ； ； ； ; Among them, cI x1 represents the equivalent moment of inertia of the first particle around the x-axis, cI y1 represents the equivalent moment of inertia of the first particle around the y-axis, cI z1 represents the equivalent moment of inertia of the first particle around the z-axis, cI xn represents the equivalent moment of inertia of the nth particle around the x-axis, cI yn represents the equivalent moment of inertia of the nth particle around the y-axis, cI zn represents the equivalent moment of inertia of the nth particle around the z axis, cm x1 represents the translational mass of the first particle on the x-axis, cm y1 represents the translational mass of the first particle on the y-axis, cm z1 represents the translational mass of the first particle on the z-axis, cm xn represents the translational mass of the nth particle on the x-axis, cm yn represents the translational mass of the nth particle on the y-axis, cm zn represents the translational mass of the nth particle on the z-axis; diag represents the diagonal matrix; represents the transformation matrix that maps physical coordinates to modal coordinates, M represents the unreduced overall mass matrix of the flexible vehicle body, represents the volume integral of the entire vehicle body structure, dV represents the integral sign, R is the free volume matrix, 0 is the zero matrix of PB×L, L is the retained elastic mode order, is the boundary-induced displacement transfer matrix, PB is the number of critical degrees of freedom; ; K II represents the internal degree of freedom stiffness matrix, K IB represents the interior-boundary degree of freedom coupling stiffness matrix, represents the fixed interface modal matrix; ;M II represents the internal degree of freedom mass matrix, represents the diagonal matrix of the first K-order eigenvalues; ρ represents the density of the body material, represents the modal shape function matrix, JD represents the Jacobian matrix of rigid body motion; The vehicle stiffness matrix is ​​K ta : ； K rr represents the vehicle rigid body stiffness matrix, K mm represents the modal stiffness matrix, K rm represents the stiffness coupling matrix, K mr K rm The transposed matrix of ； ； ; T r Represents the transformation matrix from the rigid body displacement to the suspension point, K s represents the local stiffness matrix of the suspension system, represents the transformation matrix that maps physical coordinates to modal coordinates, K represents the unreduced vehicle overall stiffness matrix, T p Represents the transformation matrix from modal coordinates to suspension points; Vehicle damping matrix C ta for: ； Among them, C rr represents the damping matrix, C rm represents the damping coupling matrix, C mm represents the modal damping matrix, C mr represents the transposed matrix of the damping coupling matrix; in, ; ; ; T r Represents the transformation matrix from the rigid body displacement to the suspension point, C s represents the suspension damping, α2 and β2 are the first vehicle damping coefficient and the second vehicle damping coefficient respectively, T p represents the transformation matrix from modal coordinates to suspension points, and T represents the transposed matrix.

8. The analysis method for integrating rail vehicle dynamics and structural reliability according to claim 1 is characterized in that: Typical operating scenarios include the operating route with the maximum turning radius, the upward route with the maximum slope, the downward route with the maximum slope, and the operating route at the highest speed.

9. The analysis method for integrating rail vehicle dynamics and structural reliability according to claim 1 is characterized in that: Step 5 includes the following steps: Step 51: Input the coupling equation into the DAP-Rail software; Step 52: inputting operation scenario parameters, which include line parameters, vehicle parameters and load conditions; Step 53: The DAP-Rail software outputs a load curve corresponding to time, and extracts an equivalent load spectrum from the load curve.

10. An analysis system integrating rail vehicle dynamics and structural reliability, characterized in that: The analysis method for integrating rail vehicle dynamics and structural reliability as described in any one of claims 1 to 9 is used to generate the reliability probability of the vehicle running for a long time on the line.

Images