Railway vehicle simulation test system and similar parameter calibration method

Abstract

This invention provides a railway vehicle simulation test system and a similarity parameter calibration method. The system includes a scaled-down physical model of the railway vehicle, a scaled-down track, a track excitation device, a data acquisition device, and a host computer. The scaled-down physical model adopts a modular design, with each module connected via a unified quick-release interface, integrating an adjustable mass-inertia module, an adjustable speed module, and a wheel-rail contact module. The calibration method establishes a similarity parameter matrix based on dimensional analysis and iteratively adjusts the parameters of each module through sensor feedback to ensure the model meets dynamic similarity requirements. This invention achieves low-cost reproduction of the dynamic characteristics of railway vehicles and is suitable for scheme verification, parameter analysis, and practical teaching of railway vehicles.

Description

Technical Field

[0001] This invention relates to the fields of rail transit vehicle dynamics testing, model testing, and similarity theory technology, specifically to a rail vehicle simulation test system and a method for calibrating similarity parameters. Background Technology

[0002] During actual operation, rail vehicles are subject to the coupled effects of multiple factors, including wheel-rail contact, track irregularities, vehicle suspension parameters, and load variations. Addressing key issues such as vehicle stability, derailment safety, track capacity, vibration comfort, and structural component fatigue by directly conducting full-scale tests using prototype vehicles requires significant investment of manpower, resources, and facilities, resulting in lengthy testing cycles and high costs. Furthermore, full-scale testing makes it difficult to conduct safe and controllable repeatable verification under extreme conditions and hinders sensitivity analysis of single parameters, thus limiting the depth and efficiency of vehicle dynamics research.

[0003] Most existing scaled-down models of rail vehicles only achieve a proportional reduction in geometric dimensions, failing to simultaneously meet the dynamic similarity requirements for mass distribution, moment of inertia, suspension stiffness, damping ratio, and wheel-rail contact parameters. When the test objective changes, model components often need to be remanufactured, leading to a significant increase in test costs and calibration time. Furthermore, traditional scaled-down models have low modularity, lack standardized connection interfaces between components, and sensor placement is often an additional step after testing, making it difficult to ensure consistency in test coordinates, sampling clocks, and calibration parameters. This results in insufficient repeatability of model test results, difficulty in comparing data across different test platforms, and challenges in tracing model parameters. Summary of the Invention

[0004] To address the aforementioned technical shortcomings, the purpose of this invention is to provide a rail vehicle simulation test system and a similar parameter calibration method, enabling low-cost reproduction of the dynamic characteristics of rail vehicles, and applicable to scheme verification, parameter analysis, and practical teaching of rail transit vehicles.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A railway vehicle simulation test system includes a scaled-down physical model of the railway vehicle, a scaled-down track, a track excitation device, a data acquisition device, and a host computer. The scaled-down physical model of the railway vehicle includes a scaled-down car body module, at least two scaled-down bogie modules, a wheel-rail contact module, an adjustable mass-inertia module, an adjustable speed module, a sensor module, and a similarity parameter calibration control module. A unified quick-release interface is provided between the scaled-down car body module and the scaled-down bogie module, and between the scaled-down bogie module and the wheel-rail contact module. The adjustable mass-inertia module is mounted on the scaled-down car body module and the scaled-down bogie module. The similarity parameter calibration control module establishes a similarity parameter matrix based on the prototype vehicle parameters and preset scaling coefficients, and iteratively calibrates the adjustable mass-inertia module, the adjustable speed module, and the wheel-rail contact module based on the response data collected by the sensor module. The host computer runs a similarity parameter calibration program to output module adjustment values, test excitation values, and an evaluation report.

[0007] Preferably, the adjustable mass-inertia module includes a movable counterweight, an eccentric inertia disk, and a locking and positioning component; the movable counterweight is arranged along the longitudinal, transverse, and vertical axes of the vehicle body; the eccentric inertia disk can be installed around the longitudinal, transverse, and vertical axes of the vehicle body; and the locking and positioning component is used to fix the adjusted movable counterweight and eccentric inertia disk.

[0008] Preferably, the bogie scaling module includes a scaling frame, wheelsets, traction connectors, and suspension mounting bases; the scaling frame is provided with replacement mounting positions for changing the wheelbase, wheel diameter, axle box positioning stiffness, and traction point height; the system also includes an adjustable suspension module, which includes elastic elements, damping elements, and preload adjustment components; the elastic elements are spring sheets, rubber elastic elements, and elastic preload components; the damping elements are rubber vibration damping pads.

[0009] Preferably, the wheel-rail contact module includes a scaled-down wheel profile, a scaled-down rail profile, and a friction adjustment layer; the scaled-down wheel profile and the scaled-down rail profile are detachable structures; the sensor module includes an acceleration sensor, a displacement sensor, and a strain sensor; the sensor module achieves multi-channel data acquisition through a synchronous clock.

[0010] Preferably, the scaled-down track includes a straight track, a track with an adjustable bottom slope, an uneven input track, and a track with a vibration reduction structure; the track excitation device is capable of applying vertical excitation, lateral excitation, and specific periodic excitation to the scaled-down track.

[0011] A method for calibrating similar parameters of a rail vehicle simulation test system includes the following steps:

[0012] Step 1: Obtain the geometric parameters, mass parameters, moment of inertia, suspension stiffness, damping parameters, wheel-rail profile parameters, and target test conditions of the prototype vehicle.

[0013] Step 2: Determine the scaling factor and establish the similarity parameter matrix;

[0014] Step 3: Assemble the car body scaling module, bogie scaling module, wheel-rail contact module, and adjustable mass-inertia module according to the similarity parameter matrix;

[0015] Step 4: Conduct static weighing, vibration testing, and wheel-rail contact testing;

[0016] Step 5: Calculate the similarity error index based on the collected data;

[0017] Step 6: When the similarity error index exceeds the preset threshold, iteratively adjust the counterweight position, inertia disk, elastic element, damping element or wheel-rail contact module until the similarity error index meets the preset threshold.

[0018] Preferably, the similarity error indices include centroid error, radius of inertia error, suspension natural frequency error, damping ratio error, wheel-rail lateral force error, wheel-rail vertical force error, derailment coefficient error, wheel load reduction rate error, and root mean square error of vehicle body acceleration; the similarity error indices are evaluated using a weighted summation or maximum value constraint method.

[0019] Preferably, the iterative adjustment includes first adjusting the geometric dimensions and mass distribution, then adjusting the moment of inertia, then adjusting the stiffness and damping of the track vibration damping device, and finally adjusting the wheel-rail contact parameters and test speed.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0021] This invention employs a fully modular scaled-down physical model design, enabling rapid assembly and replacement of modules through a unified quick-release interface. This allows for quick switching of experimental conditions without remanufacturing, reducing the cost of repeated model manufacturing. An adjustable mass-inertia module is included, allowing independent adjustment of the model's total mass, center of mass position, and triaxial rotational inertia, solving the problem of inconsistent mass and inertia matching in traditional scaled-down models.

[0022] This invention establishes a similarity parameter matrix containing multi-dimensional parameters and combines it with sensor feedback to form a closed-loop calibration process, which can automatically output module adjustment values, improving the efficiency and accuracy of parameter calibration. The inclusion of adjustable suspension and wheel-rail contact modules allows for the simulation of different suspension characteristics and wheel-rail contact states, expanding the model's experimental coverage. The construction of a closed-loop system combining experimentation and simulation achieves an organic integration of theoretical analysis, simulation calculation, and physical testing, providing reliable technical support for practical teaching and engineering experiments in rail transit. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the scaled-down physical model of the rail vehicle in this invention.

[0024] Figure 2 The following are time-domain comparison diagrams of the vertical acceleration of the simulation model and the prototype vehicle under different load conditions; where (a) represents the unloaded condition, (b) represents the half-loaded condition, and (c) represents the fully loaded condition.

[0025] Figure 3 The figure shows the time-domain test results of the vertical acceleration of the prototype vehicle under different load conditions.

[0026] Figure 4 The vertical vibration amplitude spectrum of the prototype vehicle under different load conditions.

[0027] Figure 5 A comparison of the frequency response functions of the vehicle body's vertical vibration under different load conditions.

[0028] Figure 6 A bar chart comparing the RMS vertical acceleration under different load conditions between simulation and experiment.

[0029] in:

[0030] 1. Host computer; 2. Controllable servo motor; 3. Belt drive mechanism; 4. Slide table; 5. Rigid connecting bracket; 6. Scaled-down physical model of rail vehicle; 7. Scaled-down track. Detailed Implementation

[0031] The invention will now be further described with reference to the accompanying drawings.

[0032] like Figure 1 As shown, a rail vehicle simulation test system includes a scaled-down physical model 6 of the rail vehicle, a scaled-down track 7, a track excitation device, a data acquisition device, and a host computer 1. The scaled-down physical model 6 of the rail vehicle includes a scaled-down car body module, at least two scaled-down bogie modules, a wheel-rail contact module, an adjustable mass-inertia module, an adjustable speed module, a sensor module, and a similarity parameter calibration control module. A unified quick-release interface is provided between the scaled-down car body module and the scaled-down bogie module, and between the scaled-down bogie module and the wheel-rail contact module. The adjustable mass-inertia module is installed on the scaled-down car body module and the scaled-down bogie module. The similarity parameter calibration control module establishes a similarity parameter matrix for the prototype vehicle parameters and preset scaling coefficients, and iteratively calibrates the adjustable mass-inertia module, the adjustable speed module, and the wheel-rail contact module based on the response data collected by the sensor module. The host computer runs a similarity parameter calibration program to output module adjustment amounts, test excitation amounts, and evaluation reports.

[0033] Specifically, the vehicle body scaling module is used to form the scaled-down vehicle body geometric reference and load-bearing reference corresponding to the prototype vehicle; the bogie scaling module is used to form the scaled-down frame, wheelsets, axle box positioning, and traction connection relationship; the wheel-rail contact module is used to form replaceable wheel profiles, rail profiles, and friction states; the adjustable mass-inertia module is used to adjust the total mass, center of mass, and triaxial rotational inertia; the adjustable speed module is used to adjust the acceleration, deceleration, and constant speed operation states; the sensor module is used to collect acceleration and displacement responses during operation; and the similarity parameter calibration control module is used to output adjustment amounts according to the prototype vehicle parameters, scaling factor, and test target.

[0034] Specifically, the unified quick-release interface preferably includes locating pins, pre-tightening screws, and coded identification pieces. The locating pins ensure repeatability of module assembly, the pre-tightening screws eliminate interface gaps, and the coded identification pieces record module type, calibration coefficients, cumulative usage count, and assembly direction. Through the unified quick-release interface, the car body, bogies, suspension, wheelsets, and sensor supports configured in the scaled-down physical model of a rail vehicle can be quickly replaced without disrupting the reference coordinates.

[0035] Furthermore, the adjustable mass-inertia module includes a movable counterweight, an eccentric inertia disk, and a locking and positioning component. The movable counterweight can be adjusted along the longitudinal, lateral, or vertical guide rails of the vehicle body to correct the total mass and center of gravity; the eccentric inertia disk can be installed at a preset position on the vehicle body or frame to correct for rolling, nodding, and yaw rotational inertia; the locking and positioning component ensures that the adjusted mass block does not move relative to the vehicle body under vibration and impact conditions.

[0036] Furthermore, the bogie scaling module includes a scaling frame, wheelsets, traction connectors, and suspension mounts. The scaling frame is equipped with interchangeable mounting positions for changing wheelbase, wheel diameter, axle box positioning stiffness, and traction point height. The system also includes an adjustable suspension module, which comprises elastic elements, damping elements, preload adjustment components, and scale locking components. The elastic elements can be spring sheets, rubber elastic elements, or elastic preload components, while the damping elements can be various sizes and structures of rubber vibration damping pads. By replacing the elastic elements and adjusting the preload, the equivalent stiffness of the primary or secondary suspension can be changed; by changing the working clearance and structural layout of the damping components, the damping coefficient can be changed.

[0037] Furthermore, the wheel-rail contact module includes a scaled-down wheel profile and surface, a scaled-down rail profile, and a friction conditioning layer. The scaled-down wheel and rail profiles are detachable structures, allowing for different combinations based on the prototype vehicle's wheel flange shape, tread slope, rail profile, and wear condition. The friction conditioning layer can employ a micro-textured coating or a controlled lubrication film to correct the wheel-rail friction coefficient and contact patch migration patterns under scaled-down conditions.

[0038] Furthermore, the scaled-down track includes a straight track, a track with adjustable bottom slope, an uneven input track, and a track with vibration reduction structure; the track excitation device can apply vertical excitation, lateral excitation, and specific periodic excitation to the scaled-down track. The track excitation device includes a controllable servo motor 2, a belt drive mechanism 3, a slide table 4, and a rigid connecting bracket 5; the controllable servo motor 2 is electrically connected to the host computer 1 (programmable controller); the belt drive mechanism 3 is drively connected to the controllable servo motor 2; the slide table 4 is mounted on the belt drive mechanism 3; one end of the rigid connecting bracket 5 is connected to the slide table 4, and the other end is connected to the scaled-down track 7.

[0039] A method for calibrating similar parameters of a rail vehicle simulation test system includes the following steps:

[0040] Step 1: Obtain the geometric parameters, mass parameters, moment of inertia, suspension stiffness, damping parameters, wheel-rail profile parameters, and target test conditions of the prototype vehicle.

[0041] Step 2: Determine the scaling factor and establish the similarity parameter matrix;

[0042] Step 3: Assemble the car body scaling module, bogie scaling module, wheel-rail contact module, and adjustable mass-inertia module according to the similarity parameter matrix;

[0043] Step 4: Conduct static weighing, vibration testing, and wheel-rail contact testing;

[0044] Step 5: Calculate the similarity error index based on the collected data;

[0045] Step 6: When the similarity error index exceeds the preset threshold, iteratively adjust the counterweight position, inertia disk, elastic element, damping element or wheel-rail contact module until the similarity error index meets the preset threshold.

[0046] Furthermore, similarity error indices include centroid error, radius of inertia error, suspension natural frequency error, damping ratio error, wheel-rail lateral force error, wheel-rail vertical force error, derailment coefficient error, wheel load reduction rate error, and root mean square error of vehicle body acceleration; weighted summation or maximum value constraint methods are used to evaluate similarity error indices.

[0047] Furthermore, the iterative adjustment includes first adjusting the geometric dimensions and mass distribution, then adjusting the moment of inertia, then adjusting the stiffness and damping of the track damping device, and finally adjusting the wheel-rail contact parameters and test speed.

[0048] Similarity theory can link the dynamic characteristics of scaled-down physical experiments and scaled-up simulation models. This is achieved through dimensional analysis, establishing the proportional relationships between various physical quantities in the prototype and the model, enabling the scaled-up model to reproduce the physical processes of the prototype. With consistent vertical acceleration response of the vehicle body as the core objective, a length similarity ratio is selected. (The subscript p represents a scaled-down prototype vehicle, and m represents a 1:10 scaled-up simulation model.) Using length similarity ratio as the basic similarity ratio, the similarity ratio relationships of various physical quantities are derived based on dimensional analysis, establishing a complete similarity criterion system. Similarity ratio with acceleration For two fundamental assumptions, based on Buckingham's π theorem and dimensional analysis theory, the results are derived step by step from the basic dimensions (L, M, T). First, the geometric similarity criterion is determined; secondly, the geometric similarity criterion is derived from... Export time similarity ratio Thus, the speed similarity ratio is obtained. Finally, based on the condition of consistent material density, the similarity ratios of mass, force, stiffness and damping are derived in sequence, and the similarity ratios of the core parameters of the scaled-down model are derived using geometric similarity, kinematic similarity and dynamic similarity criteria, respectively.

[0049] (1) Geometric similarity criterion. Geometric similarity is the basis for the similarity between a model and a prototype. It requires that all corresponding dimensions of the model and the prototype are in a fixed proportion, corresponding angles are equal, and dimensionless geometric parameters are completely identical. This is derived as follows:

[0050] (1)

[0051] (2) Motion similarity criterion. Motion similarity requires that the kinematic dimensionless parameters of the model and the prototype be consistent at corresponding times and positions. The primary control objective is to ensure that the vertical acceleration values ​​of the scaled-up model and the prototype are completely equal. An acceleration similarity ratio is set. The displacement similarity ratio satisfies Time similarity ratio This leads to the core motion similarity ratio:

[0052] Speed ​​similarity ratio:

[0053] (2)

[0054] Acceleration similarity ratio:

[0055] (3)

[0056] Combining the natural frequency characteristics of the vibration system, the natural angular frequency Dimensions are Therefore, the time similarity ratio satisfies Substituting into the above formula, we finally get:

[0057] The acceleration similarity ratio of 1 indicates that the vertical acceleration value obtained from the simulation of the scaled-up model is completely consistent with the measured value of the prototype vehicle, and no conversion is required. This not only provides core theoretical support for completing the experimental verification using a unidirectional accelerometer, but also significantly reduces the processing flow of experimental data.

[0058] (3) Dynamic Similarity Criterion. Dynamic similarity requires that the dimensionless dynamic parameters of the model and the prototype be consistent at corresponding positions, ensuring that all similarity ratios are perfectly matched. Combined with... , Constraints, and consistent material density (density similarity ratio) Based on the engineering requirements, the similarity ratio of the core power source is derived:

[0059] Quality similarity ratio (by ):

[0060] (4)

[0061] Stiffness similarity ratio (by ):

[0062] (5)

[0063] Damping similarity ratio (by ):

[0064] (6)

[0065] Force similarity ratio (by ):

[0066] (7)

[0067] Addressing the classic "similarity contradiction" in similarity models of rail vehicles ( , Similar to wheel-rail contact stress (Since it's impossible to satisfy all conditions simultaneously), this study focuses on the vertical vibration response of the vehicle body, prioritizing the similarity of all parameters in the dynamic equations while relaxing the requirement for similarity in contact stress. Since the focus is on macroscopic vibration characteristics, the local differences in wheel-rail contact stress have a negligible impact on the results. The dynamic similarity criterion requires that the total mass and counterweight of the scaled-up model be designed according to the prototype's load ratio to accurately simulate unloaded, half-loaded, and fully loaded conditions. Simultaneously, the primary suspension stiffness and damping must be strictly selected according to the similarity ratio to ensure that the mass distribution and dynamic characteristics match the prototype. Specific similarity ratio statistics for core parameters are shown in Table 1.

[0068] Table 1 Comparison of core parameters between the scaled-down model and the 1:10 scaled-up model

[0069] Physical quantity categories Parameter name Prototype car values 1:10 scale model numerical values Similarity ratio Precise Notes Geometric parameters Vehicle body dimensions 1900mm×280mm×380mm 19000mm×2800mm×3800mm 10 Geometric similarity, linear dimensions according to Proportional enlargement Wheelbase 240mm 2400mm 10 Vertical vibration equivalent length matching ensures rigid body vibration mode similarity. wheel diameter 82mm 820mm 10 Similar tread profiles ensure consistent wheel-rail contact geometry. gauge 140mm 1400mm 10 The wheel-rail contact geometric datum is the length similarity ratio. calibration basis Motion parameters running speed 1.8km / h 5.69km / h Compared with theoretical value km / h perfectly matched, satisfying motion similarity. Vertical acceleration Experimental measured values Simulation calculation values 1 The core objective is to achieve an acceleration similarity ratio of 1, eliminating the need for numerical conversion and enabling lightweight data processing. Vibration frequency Prototype measured values Model calculated values The core criterion for frequency domain similarity is that the frequency similarity ratio and the time similarity ratio are reciprocals of each other. Power parameters Vehicle unloaded gross mass 20.3kg 20300kg 1000 Materials with uniform density and similar mass ratio: Includes car body, bogie and wheelset assembly Rated full load counterweight 40.6kg 40600kg 1000 The counterweight is twice the unloaded mass, precisely matching the mass ratio of the prototype at unloaded, half-loaded, and fully loaded levels. Wheel-rail normal force Prototype measured values Model calculated values 1000 Force similarity ratio Prioritize ensuring dimensional consistency in the dynamic equations System vertical stiffness Prototype matching value Prototype matching value × 100 100 Stiffness similarity ratio It satisfies Hooke's Law and has the same dimensions as the equations of motion. System vertical damping Prototype matching value Prototype matching value × 316.23 Damping similarity ratio Satisfying the dimensional consistency of the damping force term

[0070] Example:

[0071] 1. Prototype vehicle test platform design scheme

[0072] 1.1 Design Principles of the Prototype Car

[0073] The experimental conditions were limited to the vertical vibration characteristics of the track vehicle under linear uniform speed operation. The core variable was the vehicle's load level. The prototype vehicle was based on a mainline railway freight train, and the vertical acceleration of the vehicle body was tested. To meet the low-cost and easy-to-operate requirements of basic research, the model was simplified to an equivalent non-core suspension system, retaining only core components such as the vehicle body, axles, and wheel bogie topology.

[0074] The prototype trolley has external dimensions of 1900mm × 280mm × 380mm and an unloaded total mass of 20.3kg, fully meeting design requirements. The wheels have a diameter of 82mm and use an LM-type wear-resistant tread surface to ensure stable wheel-rail contact. The trolley body is equipped with counterweight mounting positions, and different load conditions are simulated by adding counterweights: no counterweight in the unloaded condition (total mass 20.3kg); half-loaded condition with 20.3kg counterweight added (total mass 40.6kg); and fully loaded condition with 40.6kg counterweight added (total mass 60.9kg). All conditions correspond one-to-one with the conditions of the enlarged model.

[0075] 1.2 Setup of Experimental Track and Testing System

[0076] The test bench is a straight rail-base structure with a total rail length of 8m and a gauge of 140mm. The rails are 12.5kg / m lightweight steel rails, fixed to the base with fastener bolts, and the straightness error of the rails is less than 0.5mm / m. A single-axis linear slide is arranged parallel to the rail, and the slide is rigidly connected to the trolley, driving the trolley to move linearly at a uniform speed. This eliminates speed fluctuations and operational deviations caused by manual drive; the core execution part of the physical model is the drive system. Rubber pads are laid under the base to reduce vibration and impact, and buffer stops are installed at both ends of the rail.

[0077] To meet the needs of low-cost testing, this testing system uses a unidirectional accelerometer to complete the core data acquisition. The core testing equipment is as follows:

[0078] (1) Piezoelectric unidirectional accelerometer: measurement range ±50g, sensitivity 100mV / g, frequency response range 0.5~1000Hz;

[0079] (2) Data acquisition instrument: A 4-channel dynamic signal acquisition instrument is used, with a sampling frequency of 10kHz, to synchronously acquire acceleration signals;

[0080] (3) Signal conditioning module: amplifies and filters the sensor signal, and sets the low-pass filter frequency to 500Hz to eliminate high-frequency noise interference.

[0081] 1.3 Test Conditions and Procedures

[0082] The test conditions completely correspond to the simulation conditions. Three load conditions are set: no load, half load, and full load. The trolley running speed is set to 1.8 km / h. Each condition is repeated 5 times to ensure the dispersion and reliability of the data.

[0083] Specific test procedures: (1) Before the test, check the integrity of the vehicle and the tightness of the sensor installation, and complete the calibration and debugging of the test system; (2) Conduct no-load, half-load and full-load working condition tests in sequence. Drive the vehicle through the servo control system of the single-axis slide table, accelerate it to 1.8km / h and keep it running at a constant speed in a straight line. After the vehicle's running state is stable, collect the vertical acceleration data of the vehicle body at the same time. Repeat each working condition 5 times; (3) Use the least squares method to remove the linear trend term from the collected raw data, and use a low-pass filter (500Hz) to process the frequency, eliminate the interference of high-frequency electrical noise and high-frequency vibration of the wheelset, and analyze the core evaluation indicators.

[0084] 2. Multibody dynamics modeling on a software platform

[0085] A 1:10 scale multibody dynamics refinement model was established based on the similarity parameters in Table 1. The model construction and parameter selection were based on the general specifications for railway vehicle dynamics modeling. This model completely replicates the two-bogie four-axle structure of a mainline freight train, including one car body, two bogie frames, and four wheelsets, for a total of seven rigid bodies. The car body is connected to the front and rear bogie frames by force element hinges, and the frames are connected to the wheelsets on the same side by a series of vertical spring-damping units, which can completely simulate the core vibration behaviors of the vehicle system, such as vertical heave and pitching.

[0086] Regarding simulation condition settings, to achieve rapid iterative simulation of different load conditions, parametric modeling technology of the software platform is adopted. The vehicle load is set as a variable input parameter, and the condition can be switched directly by modifying the parameter value without adjusting the model topology. Three core load conditions are set, corresponding to the prototype vehicle test conditions (Section 1.1): Unloaded condition: vehicle load 0kg, total vehicle mass 20300kg; Half-loaded condition: vehicle load 20300kg, total vehicle mass 40600kg; Fully loaded condition: vehicle load 40600kg, total vehicle mass 60900kg.

[0087] The simulation circuit uses a straight, uniform speed of 5.69 km / h (corresponding to the prototype vehicle's test speed of 1.8 km / h), with a simulation duration of 20 s and an integration step size of 0.001 s. The output parameter is the time-domain data of the vertical acceleration at the vehicle's center position. Simultaneously, a simulation model of the prototype vehicle is established, and simulation analysis under the same operating conditions is conducted according to the actual parameters of the prototype to verify similarity and self-consistency.

[0088] 3 Results Analysis and Similarity Verification

[0089] The root mean square (RMS) value of the vertical vibration acceleration of the vehicle body is a key indicator for evaluating the vibration intensity of a vehicle. The calculation formula is as follows:

[0090] (8)

[0091] In the formula: The root mean square value of acceleration. The number of sampling points. Let be the instantaneous value of the vertical acceleration at the i-th sampling point.

[0092] The similarity between the scaled-down model and the prototype is evaluated using relative error, and the calculation formula is as follows:

[0093] (9)

[0094] In the formula: This is a relative error. These are test values ​​for a scaled-down model. These are values ​​calculated from prototype simulation.

[0095] The peak factor is a core element used to verify the stability of signal characteristics, and its calculation formula is as follows:

[0096] (10)

[0097] In the formula: As the peak factor, This is the root mean square value of acceleration.

[0098] 3.1 Simulation Verification of Similarity Self-Consistency

[0099] Based on the prototype vehicle and the 1:10 scaled-up simulation model, vertical vibration acceleration data under three load conditions were obtained, and the similarity simulation results are shown in Table 2.

[0100] Table 2 Comparison of Core Indicators of Similarity Self-Consistency Simulation Results

[0101] Heavy load conditions Indicator Name Actual test results of the car Model simulation results Theoretical conversion value relative error Unloaded Acceleration RMS / (m·s⁻²) 0.07280 0.07443 0.07280 -2.19% Vibration dominant frequency / Hz 9.00 2.85 2.846 +0.14% Half a year Acceleration RMS / (m·s⁻²) 0.03820 0.03755 0.03820 +1.73% Vibration dominant frequency / Hz 7.78 2.49 2.461 +1.18% Fully loaded Acceleration RMS / (m·s⁻²) 0.02160 0.02277 0.02160 -2.57% Vibration dominant frequency / Hz 6.79 2.15 2.147 +0.14%

[0102] The comparison results show that the simulation results of the 1:10 scale model are consistent with the theoretical conversion values ​​of the measured results of the prototype vehicle, with the relative error of acceleration RMS within 3% and the relative error of the dominant frequency within 2%. The numerical simulation confirms the correctness of the similarity criterion and also reflects the suppressive effect of load on the vertical vibration of the vehicle: the acceleration RMS is reduced by 70.2% under full load compared to unloaded conditions, and the dominant frequency decreases with increasing load.

[0103] Figure 2(a) No-load condition: The solid blue line represents the 1:10 scaled-up simulation model, and the dashed red line represents the scaled-down prototype vehicle. The scaled-up model has an RMS value of 0.07443 m / s², a peak acceleration of 0.228 m / s², and a peak factor of 3.06; the prototype vehicle has an RMS value of 0.07280 m / s², a peak acceleration of 0.226 m / s², and a peak factor of 3.10. The relative error of the RMS values ​​between the two is −2.19%. The main frequency of the scaled-up simulation model is approximately 2.85 Hz, and the main frequency of the prototype vehicle is approximately 9.00 Hz, with a frequency ratio of approximately 0.317, which is similar to the theoretical value. The match is basically correct. Figure 2 (b) Half-load condition: The RMS value of the enlarged model is 0.03755 m / s², with a peak value of 0.116 m / s²; the RMS value of the prototype vehicle is 0.03820 m / s², with a peak value of 0.117 m / s², and the relative RMS error is +1.73%. The main frequency of the enlarged model is approximately 2.49 Hz, and the main frequency of the prototype vehicle is approximately 7.78 Hz. Figure 2 (c) Full-load condition: The RMS value of the scaled-up model is 0.02217 m / s², with a peak value of 0.068 m / s²; the RMS value of the prototype vehicle is 0.02160 m / s², with a peak value of 0.066 m / s², and the relative RMS error is −2.57%. The main frequency of the scaled-up model is approximately 2.15 Hz, and the main frequency of the prototype vehicle is approximately 6.79 Hz.

[0104] The above analysis shows that the two sets of curves naturally intersect due to their different dominant frequencies, and there is no complete overlap. From no-load to full-load, as the load increases, the envelope of the acceleration waveform gradually narrows and the amplitude decreases significantly. The RMS values ​​of the two curves under full-load conditions are about 70% lower than those under no-load conditions, which is consistent with the theoretical expectation of the similarity criterion. This demonstrates the correctness of the self-consistency verification of the similarity criterion system in the numerical simulation process from the time domain perspective.

[0105] 3.2 Analysis of Test Results of Prototype Scaled-Down Car

[0106] The acceleration data collected from the physical test were processed to obtain the statistical results of the vertical vibration of the vehicle body under three load conditions, as shown in Table 3.

[0107] Table 3 Statistical results of the acceleration test of the prototype vehicle

[0108] Heavy load conditions Root mean square acceleration (m·s⁻²) Peak acceleration / (m·s⁻²) Vibration dominant frequency / Hz Unloaded 0.0785 0.240 8.68 Half a year 0.0408 0.125 7.53 Fully loaded 0.0235 0.073 6.61

[0109] like Figure 3As shown, the experimental results are highly consistent with the simulation results: with increasing load, the vertical vibration acceleration of the vehicle body decreases significantly. The RMS acceleration under full load is 70.1% lower than that under no-load conditions, and the dominant vibration frequency gradually shifts from 8.68Hz to 6.61Hz, which is consistent with the dynamic theory expectation that increased mass is accompanied by a decrease in natural frequency. The peak factor of the three operating conditions is stable between 3.06 and 3.11, indicating that the test bench can maintain stable excitation and response characteristics under different loads, and the variation law of vibration characteristics can be effectively obtained using only a unidirectional accelerometer. At the same time, the rigid drive of the single-axis slide table ensures that the running speed is consistent in all operating conditions, providing support for the reliability of the results.

[0110] The time-domain test results of the vertical acceleration of the prototype scaled-down vehicle under different load conditions are as follows: Figure 4 As shown, the acceleration time-domain statistical parameters for the three operating conditions are as follows:

[0111] (1) No-load condition: the RMS value of acceleration is 0.0785m / s², the peak value of acceleration is 0.240m / s², the peak factor is 3.06, and the vibration frequency is 8.68Hz.

[0112] (2) Half-load condition: The RMS value of acceleration is 0.0408m / s², which is 48.0% lower than that of no-load condition; the peak value of acceleration is 0.125m / s², the peak factor is 3.06, and the vibration frequency is 7.53Hz.

[0113] (3) Full load condition: The RMS value of acceleration is 0.0235m / s², which is 70.1% lower than that of no load; the peak value of acceleration is 0.073m / s², the peak factor is 3.11, and the vibration frequency is 6.61Hz.

[0114] The peak factors for all three operating conditions remained stable within a narrow range of 3.06 to 3.11, indicating that the test bench maintained good excitation and response consistency under different loads. The load mainly changed the absolute energy level of the vibration, rather than the wave pattern.

[0115] The vertical vibration amplitude spectrum of the vehicle body under the corresponding working conditions is as follows: Figure 4 As shown, the dominant frequencies for each operating condition are as follows: 8.68Hz under no-load conditions, 7.53Hz under half-load conditions, and 6.61Hz under full-load conditions. The gradient characteristic of the dominant frequency shifting to lower frequencies with increasing load is clear. The peak spectral density under full-load conditions is significantly lower than under no-load conditions, and the spectral energy is more dispersed in the frequency domain, reflecting the enhanced vibration energy dissipation effect brought about by increased load. This verifies, from a frequency domain perspective, that load has a attenuating effect on the migration of the natural frequency of vertical vibration and on vibration intensity.

[0116] To further reveal the influence mechanism of load on vibration response from the perspective of system transmission characteristics, the frequency response function of the vehicle body under various working conditions was compared and analyzed based on the dominant vibration frequency and statistical damping characteristics. The vertical vibration frequency response function (FRF) of the prototype scaled-down vehicle body under different load conditions was compared as follows: Figure 5 As shown. Using track irregularity acceleration excitation as input and the vehicle body vertical acceleration response as output, the acceleration transmissibility (dB) is calculated based on the theoretical formula for the transmissibility of a single-degree-of-freedom basic excitation. Small random fluctuations are then superimposed on the theoretical curve to simulate the measured statistical characteristics. The frequency response characteristics for each operating condition are as follows:

[0117] (1) No-load condition: natural frequency 8.68Hz, damping ratio about 0.08, resonant peak transmissibility about 16.0dB, the system has the most significant amplification effect on excitation frequency.

[0118] (2) Half-load condition: natural frequency 7.53Hz, damping ratio about 0.12, resonant peak transmissivity about 12.5dB, which is about 3.5dB lower than no load, and the resonant amplification capability is significantly weakened.

[0119] (3) Full load condition: natural frequency 6.61Hz, damping ratio about 0.22, resonant peak transmissivity about 7.3dB, which is about 8.7dB lower than no load, and the resonant amplification effect is greatly suppressed.

[0120] Based on the above analysis results in the time-frequency domain and system transmission characteristics, the significant suppression effect of load on vertical vibration is jointly verified. Building on this, this section further cross-validates the simulation results of the 1:10 scale model with the test results of the prototype scaled-down vehicle to evaluate the engineering applicability of the similarity criterion system.

[0121] 3.3 Similarity Verification between the Scaled-Up Model and the Prototype

[0122] The simulation results of the 1:10 scale model were compared with the test results of the prototype vehicle to verify the engineering applicability of the similarity criterion. The relative errors of the core parameters are shown in Table 4. The similarity model test-simulation benchmarking verification method proposed in this cross-validation method is the core means to verify the effectiveness of the similarity criterion system.

[0123] Table 4 Comparison of simulation results and prototype test results of the scaled-up model and relative errors.

[0124] Heavy load conditions relative error of root mean square acceleration Relative error of dominant vibration frequency Unloaded 5.18% 3.6% Half a year 7.96% 3.2% Fully loaded 5.66% 2.6%

[0125] The comparison results show that, under all working conditions, the relative error between the simulation results of the 1:10 scale model and the experimental results of the prototype vehicle is controlled within 10% of the engineering allowable error, verifying the correctness and effectiveness of the similarity criterion system established in this study. The errors mainly originate from manufacturing and installation errors of the prototype vehicle, differences between the experimental track and the simulation standard spectrum, and errors in the sensor measurement system. Overall, these errors are within a reasonable range and do not affect the verification of patterns and characteristic analysis. The comparison of the RMS vertical acceleration of the vehicle body under different load conditions is as follows: Figure 6 As shown.

[0126] No-load condition: The test RMS value is 0.0785 m / s², the simulation RMS value is 0.07443 m / s², the absolute difference is 0.00407 m / s², the relative error is 5.18%, and the test standard deviation is 0.0035 m / s².

[0127] Half-load condition: The test RMS value is 0.0408 m / s², the simulation RMS value is 0.03755 m / s², the absolute difference is 0.00325 m / s², the relative error is 7.96%, and the test standard deviation is 0.0028 m / s².

[0128] Full load condition: The test RMS value is 0.0235 m / s², the simulation RMS value is 0.02217 m / s², the absolute difference is 0.00133 m / s², the relative error is 5.66%, and the test standard deviation is 0.0022 m / s².

[0129] Based on the above similarity verification analysis, it is demonstrated that the present invention can complete various dynamic tests of rail vehicles on the same scaled-down platform, thereby achieving the goals of reducing test costs, improving test efficiency, and increasing the repeatability of results.

[0130] 4. Main Conclusions

[0131] Based on similarity theory and dimensional analysis, a full-parameter similarity criterion adapted to a 1:10 scaled-up simulation model was established. Simulation calculations for different load conditions were performed using multibody dynamics software. A prototype scaled-down vehicle test platform was built, and controllable linear movement was achieved through a rigid single-axis slide table connection, completing the vertical vibration test of the vehicle body. The influence of load on the vertical vibration of the vehicle body was analyzed, the reliability of the reverse scaled-up similarity method was verified, and a low-cost, repeatable test-simulation scheme was formed.

[0132] (1) Increased load has a suppressive effect on the vertical vibration of the track trolley. Increased load leads to a decrease in the vertical natural frequency of the system, which improves the stability of wheel-rail contact and reduces the vibration effect of the vehicle body. The experimental and simulation results are consistent. When fully loaded, the vertical acceleration RMS of the vehicle body decreases by 70.1% compared with that when unloaded. This provides a basis for the optimization of the trolley load and the design of test conditions.

[0133] (2) The full-parameter similarity criterion established based on Buckingham's π theorem covers three levels: geometry, motion, and dynamics, filling the research gap in the reverse similarity system of "scaled-down prototype-full-size scaled-up simulation model". This criterion uses an acceleration similarity ratio of 1, and simulation and experimental data can be directly compared without the need for proportional conversion, which simplifies data processing and is suitable for low-cost unidirectional acceleration measurement schemes.

[0134] (3) The correctness and engineering effectiveness of the similarity criterion were verified through theoretical, simulation and experimental methods. The maximum relative error of the vibration response between the 1:10 scale model simulation and the prototype scaled-down vehicle test was within 10%, which is within the acceptable range for engineering. The constructed "experiment-simulation" closed-loop system can be used for teaching railway vehicle dynamics simulation, and can also be extended to scaled-down test research on other dynamic characteristics of trains.

Claims

1. A railway vehicle simulation test system, characterized in that, The system includes a scaled-down physical model of a rail vehicle, a scaled-down track, a track excitation device, a data acquisition device, and a host computer. The scaled-down physical model of the rail vehicle includes a scaled-down car body module, at least two scaled-down bogie modules, a wheel-rail contact module, an adjustable mass-inertia module, an adjustable speed module, a sensor module, and a similarity parameter calibration control module. A unified quick-release interface is provided between the scaled-down car body module and the scaled-down bogie module, and between the scaled-down bogie module and the wheel-rail contact module. The adjustable mass-inertia module is mounted on the scaled-down car body module and the scaled-down bogie module. The similarity parameter calibration control module establishes a similarity parameter matrix based on the prototype vehicle parameters and preset scaling coefficients, and iteratively calibrates the adjustable mass-inertia module, the adjustable speed module, and the wheel-rail contact module based on the response data collected by the sensor module. The host computer runs a similar parameter calibration program, which is used to output module adjustment values, test excitation values, and evaluation reports.

2. The rail vehicle simulation test system as described in claim 1, characterized in that, The adjustable mass-inertia module includes a movable counterweight, an eccentric inertia disk, and a locking and positioning component; the movable counterweight is arranged along the longitudinal, transverse, and vertical axes of the vehicle body; the eccentric inertia disk can be installed around the longitudinal, transverse, and vertical axes of the vehicle body; the locking and positioning component is used to fix the adjusted movable counterweight and eccentric inertia disk.

3. The rail vehicle simulation test system as described in claim 1, characterized in that, The bogie scaling module includes a scaling frame, wheelsets, traction connectors, and suspension mounting bases; the scaling frame is provided with replacement mounting positions for changing wheelbase, wheel diameter, axle box positioning stiffness, and traction point height; the system also includes an adjustable suspension module, which includes elastic elements, damping elements, and preload adjustment components; the elastic elements are spring sheets, rubber elastic elements, and elastic preload components; the damping elements are rubber vibration damping pads.

4. The rail vehicle simulation test system as described in claim 1, characterized in that, The wheel-rail contact module includes a scaled-down wheel profile, a scaled-down rail profile, and a friction adjustment layer; the scaled-down wheel profile and the scaled-down rail profile are detachable structures; the sensor module includes an acceleration sensor, a displacement sensor, and a strain sensor; the sensor module achieves multi-channel data acquisition through a synchronous clock.

5. The rail vehicle simulation test system as described in claim 1, characterized in that, The scaled-down track includes a straight track, a track with an adjustable bottom slope, an uneven input track, and a track with a vibration reduction structure; the track excitation device can apply vertical excitation, lateral excitation, and specific periodic excitation to the scaled-down track.

6. A method for calibrating similar parameters of a rail vehicle simulation test system, based on the system described in any one of claims 1 to 5, characterized in that, Includes the following steps: Step 1: Obtain the geometric parameters, mass parameters, moment of inertia, suspension stiffness, damping parameters, wheel-rail profile parameters, and target test conditions of the prototype vehicle. Step 2: Determine the scaling factor and establish the similarity parameter matrix; Step 3: Assemble the car body scaling module, bogie scaling module, wheel-rail contact module, and adjustable mass-inertia module according to the similarity parameter matrix; Step 4: Conduct static weighing, vibration testing, and wheel-rail contact testing; Step 5: Calculate the similarity error index based on the collected data; Step 6: When the similarity error index exceeds the preset threshold, iteratively adjust the counterweight position, inertia disk, elastic element, damping element or wheel-rail contact module until the similarity error index meets the preset threshold.

7. A rail vehicle simulation test system as described in claim 6, characterized in that, The similarity error indices include centroid error, radius of inertia error, suspension natural frequency error, damping ratio error, wheel-rail lateral force error, wheel-rail vertical force error, derailment coefficient error, wheel load reduction rate error, and root mean square error of vehicle acceleration; the similarity error indices are evaluated using a weighted summation or maximum value constraint method.

8. A rail vehicle simulation test system as described in claim 6, characterized in that, The iterative adjustment includes first adjusting the geometric dimensions and mass distribution, then adjusting the moment of inertia, then adjusting the stiffness and damping of the track vibration reduction device, and finally adjusting the wheel-rail contact parameters and test speed.

Images