Method and system for characterizing air spring performance degradation

Abstract

The application provides a kind of air spring performance degradation characterization method and system, the method comprises: based on air suspension test system, obtains the spring stiffness physical parameter generated in the performance degradation test process of air spring;Spring stiffness physical parameter is input to air spring performance degradation model, and dynamic stiffness performance degradation data are obtained;Dynamic stiffness performance degradation data include dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data;Air spring performance degradation model includes dynamic stiffness frequency correlation model and dynamic stiffness amplitude correlation model, and dynamic stiffness frequency correlation model is used to obtain stiffness frequency correlation data according to air spring internal gas pressure variation data in spring stiffness physical parameter;Dynamic stiffness amplitude correlation model is used to obtain dynamic stiffness amplitude correlation data according to air bag degradation data in spring stiffness physical parameter.The application can provide more accurate air spring performance prediction and degradation analysis result.

Description

Technical Field

[0001] This invention relates to the field of air suspension system technology, and in particular to a method and system for characterizing the performance degradation of air springs. Background Technology

[0002] Air springs, as a core component of modern vehicle suspension systems, are widely used in commercial vehicles, buses, and rail transportation due to their excellent damping performance and flexibility in adjusting vehicle height.

[0003] However, with prolonged use and increasingly complex and variable operating environments, the performance of air springs gradually deteriorates. This degradation not only weakens the damping performance of the vehicle's suspension system and reduces overall handling stability, but may also pose a potential threat to driving safety.

[0004] Existing research on air spring performance is mainly limited to the static or steady-state properties of air springs, particularly the analysis of the degradation of their stiffness and damping performance over time. It lacks an effective theoretical framework to deeply reveal the dynamic behavior changes during performance degradation, making it difficult to detect potential hidden faults in air springs during performance decline. This hinders the provision of precise maintenance strategies and performance optimization measures for suspension systems. Therefore, there is an urgent need for a method and system for characterizing air spring performance degradation to address these issues. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method and system for characterizing the performance degradation of air springs.

[0006] This invention provides a method for characterizing the performance degradation of air springs, comprising:

[0007] Based on the air suspension testing system, the physical parameters of spring stiffness generated during the performance degradation test of the air spring under test are obtained.

[0008] The physical parameters of spring stiffness are input into the air spring performance degradation model to obtain the dynamic stiffness performance degradation data of the air spring under test.

[0009] The dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data; the air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model. The dynamic stiffness frequency correlation model is used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters; the dynamic stiffness amplitude correlation model is used to calculate the dynamic stiffness amplitude correlation data based on the air bladder degradation data in the spring stiffness physical parameters.

[0010] According to a method for characterizing the performance degradation of an air spring provided by the present invention, the air suspension testing system includes a hydraulic excitation device, a force sensor, and a pressure sensor. The hydraulic excitation device is positioned directly below the air spring under test and is used to apply a periodic force to the air spring under test according to preset air spring test conditions to control the compression and extension states of the air spring. The preset air spring test conditions include a preset hydraulic excitation frequency, a preset number of fatigue test cycles, and a preset air spring test amplitude. The force sensor is positioned directly above the air spring under test and is used to measure the changes in the air spring under test when subjected to force.

[0011] The air suspension testing system acquires the physical parameters of the spring stiffness generated during the performance degradation test of the air spring under test, including:

[0012] When the air spring under test is subjected to the periodic force applied by the hydraulic excitation device, the internal gas pressure data, internal gas temperature data, and effective area of ​​the internal gas of the air spring are collected as the internal gas of the air spring changes with the force.

[0013] Based on the internal gas pressure data, internal gas temperature data, and effective area of ​​the internal gas in the air spring, the internal gas pressure change data of the air spring is obtained.

[0014] Obtain the physical parameters of the air bladder in the air spring to be tested;

[0015] When the air spring under test is subjected to the periodic force applied by the hydraulic excitation device, the airbag stress relaxation data of the airbag is obtained, wherein the airbag stress relaxation data is calculated based on the airbag oxidation rate or elastic modulus change coefficient corresponding to the airbag thickness during the performance degradation test.

[0016] Based on the physical parameters of the airbag and the stress relaxation data of the airbag, the degradation data of the airbag is obtained.

[0017] The physical parameters of the spring stiffness are obtained based on the data of gas pressure change inside the air spring and the data of air bladder degradation.

[0018] According to the method for characterizing the performance degradation of an air spring provided by the present invention, the airbag stress relaxation data of the airbag is obtained through the following steps:

[0019] Based on the Arrhenius model, the coefficient of change of elastic modulus is obtained according to the thermal acceleration coefficient and the thickness of the airbag during the performance degradation test.

[0020] Based on the airbag thickness change status information and diffusion coefficient, the airbag oxidation rate corresponding to the airbag thickness when subjected to the force is obtained.

[0021] Based on the applied force and the elastic modulus variation coefficient, the airbag stress relaxation data corresponding to the airbag thickness when subjected to the applied force is calculated; or, based on the initial airbag stress data and initial oxidation rate corresponding to the airbag in the initial state, and the airbag oxidation rate, the airbag stress relaxation data corresponding to the airbag thickness when subjected to the applied force is calculated.

[0022] The thermal acceleration coefficient and the diffusion coefficient are obtained based on a few-shot deep learning network model trained by a deep learning network, according to the input preset air spring test conditions. The few-shot deep learning network model is trained using sample dynamic stiffness performance degradation data generated under sample air spring test conditions, as well as the thermal acceleration coefficient label and diffusion coefficient label corresponding to the sample dynamic stiffness performance degradation data.

[0023] According to the present invention, a method for characterizing the performance degradation of an air spring further includes:

[0024] Based on the residual results output by the few-sample deep learning network model, the dynamic stiffness performance degradation data is corrected to obtain the corrected dynamic stiffness performance degradation data.

[0025] According to the method for characterizing the performance degradation of an air spring provided by the present invention, the formula of the dynamic stiffness frequency correlation model is specifically as follows:

[0026] ;

[0027] ;

[0028] in, Indicates angular frequency as Dynamic stiffness frequency correlation data; This indicates the stiffness generated by the high-pressure gas inside the air spring under test; This refers to the equivalent damping generated by the heat exchange between the internal gas of the airbag and the gas inside the auxiliary air chamber and the outside environment. This represents the equivalent stiffness generated by the additional air chamber; This represents the equivalent damping generated by the high-pressure gas flowing through the connecting hole inside the air spring under test; Indicates specific heat ratio. This indicates the initial pressure of the gas inside the air spring being tested. This indicates the effective area of ​​the gas inside the air spring. This indicates the initial volume of the gas inside the air spring being tested. Indicates constant-volume heat capacity. This indicates the mass of gas inside the air spring being tested. Indicates the equivalent heat transfer coefficient; This indicates that the air spring under test is subjected to the force described above, and the thickness of the air bladder is... The volume of the internal gas at that time; Represents the parameters of an ideal gas. This represents the correction factor for the equivalent area of ​​the throttling orifice. This indicates the initial temperature of the gas inside the air spring being tested. This represents the equivalent damping coefficient of the throttling orifice.

[0029] According to the method for characterizing the performance degradation of an air spring provided by the present invention, the formula of the dynamic stiffness amplitude correlation model is specifically as follows:

[0030] ;

[0031] in, Indicates the test amplitude is Dynamic stiffness amplitude correlation data at time; Indicates the effective radius of the air spring. This represents the correction coefficient for the correlation between dynamic stiffness and amplitude. This indicates the thickness of the reinforcing composite material inside the air spring under test. This indicates the volume fraction of the cord in the air bladder of the air spring under test; This indicates the Young's modulus or damage modulus of the cord in the air spring under test. This represents the airbag stress relaxation data; Indicates the inclination angle of the cord. This indicates the characteristic amplitude of the airbag. This indicates the type of rubber material used in the airbag. This indicates the total thickness of the air bladder in the air spring under test. This indicates the deflection radius of the air spring under test. This indicates the piston tilt angle.

[0032] According to the method for characterizing the performance degradation of an air spring provided by the present invention, the airbag stress relaxation data is calculated by the following formula:

[0033] ;

[0034] The formula for calculating the oxidation rate of the airbag is:

[0035] ;

[0036] The formula for calculating the coefficient of change of the elastic modulus is:

[0037] ;

[0038] in, This indicates that the airbag has a thickness of [missing information]. The corresponding airbag stress relaxation data at that time, This indicates that the airbag has a thickness of [missing information]. The magnitude of the force applied at that time, This indicates the preset air spring test amplitude. This represents the coefficient of variation of the elastic modulus. This represents the initial airbag stress data. This represents the initial oxidation rate; This indicates the thickness displacement of the air spring under test before it is subjected to the applied force. This indicates the thickness displacement of the air spring under test after being subjected to the applied force; This represents the diffusion coefficient. Represents the diffusion function. Indicates the integrating factor. Indicates the gas compressibility factor. Indicates the acceleration coefficient. This indicates the change in airbag thickness. This represents the thermal acceleration coefficient. Indicates the thermal influence factor. This indicates the equivalent activation energy of the airbag. Represents the gas constant. This indicates absolute temperature.

[0039] The present invention also provides an air spring performance degradation characterization system, comprising:

[0040] The test parameter acquisition module is used to acquire the physical parameters of spring stiffness generated by the air spring under test during the performance degradation test based on the air suspension test system.

[0041] The performance degradation characterization module is used to input the physical parameters of the spring stiffness into the air spring performance degradation model to obtain the dynamic stiffness performance degradation data of the air spring under test.

[0042] The dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data; the air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model. The dynamic stiffness frequency correlation model is used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters; the dynamic stiffness amplitude correlation model is used to calculate the dynamic stiffness amplitude correlation data based on the air bladder degradation data in the spring stiffness physical parameters.

[0043] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the air spring performance degradation characterization method as described above.

[0044] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the air spring performance degradation characterization method as described above.

[0045] The air spring performance degradation characterization method and system provided by this invention combine the dynamic changes of internal gas pressure in the air spring, the nonlinear transmission characteristics of the rubber airbag, and the air spring performance degradation phenomenon caused by multiple internal and external factors, to provide more accurate air spring performance prediction and degradation analysis results. It is not only applicable to the dynamic analysis of air springs, but also provides theoretical basis and technical support for the design and control optimization of vehicle suspension systems. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0047] Figure 1 A schematic flowchart of the air spring performance degradation characterization method provided by the present invention;

[0048] Figure 2 This is a test schematic diagram of the air suspension testing system provided by the present invention;

[0049] Figure 3 A schematic diagram illustrating the working principle of the constant mass air spring provided by the present invention;

[0050] Figure 4 A schematic diagram illustrating the construction process of the few-shot deep learning network provided by this invention;

[0051] Figure 5 A schematic diagram of the air spring performance degradation characterization method provided by the present invention;

[0052] Figure 6 A schematic diagram showing the comparison between the performance degradation test results and theoretical values ​​of the air spring provided by the present invention;

[0053] Figure 7 A schematic diagram of the air spring performance degradation characterization system provided by the present invention;

[0054] Figure 8 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0056] Air springs, as an important component of modern vehicle suspension systems, are widely used in commercial vehicles, buses, and rail transportation due to their excellent damping performance and ability to adjust vehicle height. However, with increased usage time and the complexity of the working environment, air springs gradually experience performance degradation, leading to a decrease in the damping effect of the vehicle suspension system, reduced handling performance, and potentially even endangering driving safety.

[0057] Current research on air springs mainly focuses on their static or steady-state characteristics, such as stiffness and damping degradation, but pays relatively little attention to their dynamic performance changes under actual operating conditions and the influencing factors during the degradation process. Existing technologies typically rely on empirical models or experimental data for curve fitting, which cannot effectively explain the dynamic changes during performance degradation. Furthermore, due to the coupling effects of multiple internal and external factors under complex operating conditions, existing methods struggle to accurately capture the nonlinear characteristics and transient responses of air springs during dynamic operation. These problems prevent the timely identification of potential faults during air spring performance degradation, hindering the provision of effective maintenance and performance optimization solutions for the system.

[0058] In practical applications, the performance degradation of air springs is affected by multiple internal and external factors (such as temperature, humidity, and driving environment). This invention, based on an extended physical model, analyzes the impact of these environmental factors on the dynamic stiffness of air springs. Through multi-source data fusion analysis, the superposition effect of different factors under different operating conditions is studied, and the mechanism by which temperature and pressure changes affect the performance degradation of air springs is verified through experiments.

[0059] To address the problems existing in the prior art, this invention provides an interpretable dynamic hybrid characterization method for air spring performance degradation. This method encompasses key technologies such as dynamic modeling, performance degradation analysis, and control compensation for air springs. It can be applied to vehicle suspension systems, particularly addressing air spring performance degradation caused by multiple internal and external factors under small sample data environments, thereby improving the safety, comfort, and reliability of air suspension systems. The air spring performance degradation characterization method proposed in this invention considers both the nonlinear dynamic characteristics of air springs and the global and transient nature of dynamic degradation characteristics under actual operating conditions. This improves the interpretability of air spring performance degradation while providing theoretical basis and technical support for performance prediction, fault diagnosis, and control optimization.

[0060] Figure 1 A schematic flowchart of the air spring performance degradation characterization method provided by the present invention is shown below. Figure 1 As shown, the present invention provides a method for characterizing the performance degradation of an air spring, comprising:

[0061] Step 101: Based on the air suspension testing system, obtain the physical parameters of the spring stiffness generated during the performance degradation test of the air spring under test.

[0062] Figure 2 This is a test schematic diagram of the air suspension testing system provided by the present invention, which can be referred to. Figure 2 As shown, this invention uses an air suspension testing system to conduct performance degradation tests on air springs. During the performance degradation test, various physical parameters of the air springs are collected, and the spring stiffness physical parameters are calculated based on these physical parameters using corresponding calculation formulas.

[0063] Based on the above embodiments, the air suspension testing system includes a hydraulic excitation device, a force sensor, and a pressure sensor. The hydraulic excitation device is positioned directly below the air spring under test and is used to apply a periodic force to the air spring under test according to preset air spring test conditions to control the compression and extension states of the air spring under test. The preset air spring test conditions include a preset hydraulic excitation frequency, a preset number of fatigue test cycles, and a preset air spring test amplitude. The force sensor is positioned directly above the air spring under test and is used to measure the changes in the air spring under test when subjected to force.

[0064] For specific details, please refer to Figure 2As shown, the data acquisition equipment is used to collect data from various sensors, such as force and pressure, providing raw information for subsequent analysis and processing. Force sensors measure force values ​​in the system, i.e., monitoring the force on the air spring. Pressure sensors measure pressure changes in the system, such as monitoring the operating status of the high-pressure air source. The high-pressure air source provides the required gas pressure to the air spring. Hydraulic excitation is used to control the compression and extension of the air spring. Constraints represent certain limitations or constraints in the system, such as displacement limits and pressure limits. Pressure reducing valves are used to regulate the pressure of the high-pressure air source.

[0065] In this invention, the air suspension testing system is a complex system integrating various testing devices and sensors, designed to simulate the working state of the air spring under actual operating conditions, thereby providing a comprehensive and accurate evaluation of its performance. (See reference...) Figure 2 As shown, the hydraulic excitation device is installed directly below the air spring under test. Its core function is to simulate dynamic loads during actual operation by applying periodic forces. In this invention, these forces are set based on a series of preset test conditions, specifically including: a preset hydraulic excitation frequency to simulate different frequencies of road surface unevenness encountered by a vehicle during driving; for example, high-frequency vibrations may correspond to uneven urban roads, while low-frequency vibrations may simulate slight bumps on highways; a preset number of fatigue test cycles to simulate long-term, continuous operation in order to evaluate the durability and service life of the air spring; and a preset air spring test amplitude to determine the maximum displacement of the air spring during compression and extension, which helps to understand its performance under extreme conditions.

[0066] The force sensor is mounted directly above the air spring under test to ensure direct and accurate capture of changes in the air spring under force. The primary task of the force sensor is to measure the dynamic response of the air spring when subjected to periodic forces applied by a hydraulic actuator, including measuring the magnitude, direction, and time-varying nature of the force. This data allows for analysis of the air spring's stiffness, damping characteristics, and the presence of any abnormalities or signs of failure.

[0067] Figure 3 This is a schematic diagram illustrating the working principle of the air spring with constant mass provided by the present invention, which can be referred to. Figure 3 As shown, in this invention, based on the constant-mass working principle of air springs, the performance degradation test process is completed through an air suspension testing system. The constant-mass working principle of air springs means that the mass of the gas inside the air spring remains constant during its operation. For example... Figure 3As shown, the airbag is the core component of the air spring. Made of flexible material and filled with gas, it deforms when external force is applied, absorbing and releasing energy to achieve shock absorption. When the air spring is compressed or stretched, the volume of the airbag changes; this change in volume is dV. b The initial volume of the airbag is V. bo When the volume of the airbag changes, the pressure inside the airbag also changes due to the compressibility of the gas; the change in pressure is dP. b The initial pressure of the airbag is P. bo In addition, the temperature of the gas inside the airbag may change due to compression or expansion, and the amount of temperature change is dT. b The initial temperature of the airbag is T. bo .

[0068] In this invention, the ambient pressure of the air spring is atmospheric pressure (atm). Z1 and Z2 represent the height or displacement of the air spring in a specific state, which are important parameters for evaluating the working state of the air spring. h represents the current thickness (height) of the air spring. Piston is a component in the air spring system used to control the flow and compression of gas.

[0069] The air suspension testing system acquires the physical parameters of the spring stiffness generated during the performance degradation test of the air spring under test, including:

[0070] When the air spring under test is subjected to the periodic force applied by the hydraulic excitation device, the internal gas pressure data, internal gas temperature data, and effective area of ​​the internal gas of the air spring are collected as the internal gas of the air spring changes with the force.

[0071] Based on the internal gas pressure data, internal gas temperature data, and effective area of ​​the internal gas in the air spring, the internal gas pressure change data of the air spring is obtained.

[0072] Obtain the physical parameters of the air bladder in the air spring to be tested;

[0073] When the air spring under test is subjected to the periodic force applied by the hydraulic excitation device, the airbag stress relaxation data of the airbag is obtained, wherein the airbag stress relaxation data is calculated based on the airbag oxidation rate or elastic modulus change coefficient corresponding to the airbag thickness during the performance degradation test.

[0074] Based on the physical parameters of the airbag and the stress relaxation data of the airbag, the degradation data of the airbag is obtained.

[0075] The physical parameters of the spring stiffness are obtained based on the data of gas pressure change inside the air spring and the data of air bladder degradation.

[0076] In this invention, when the air spring under test is subjected to a periodic force applied by a hydraulic excitation device, the gas inside it undergoes corresponding changes. These changes include the gas pressure, gas temperature, and effective area of ​​the gas action within the air spring. This invention uses sensors to enable the air suspension testing system to collect these gas change data in real time. A pressure sensor measures the pressure change of the internal gas, a temperature sensor monitors the temperature change, and the effective area data can be obtained through direct or indirect measurement. Based on the collected gas pressure, temperature, and effective area data, this invention allows for the calculation of specific data regarding the change in internal gas pressure of the air spring over time or with the applied force.

[0077] Furthermore, the physical parameters of the air bladder in the air spring under test are obtained, such as the initial size, shape, and material of the air bladder. Under the application of a periodic force by the hydraulic excitation device, the system monitors the stress relaxation of the air bladder. This data is calculated based on the air bladder oxidation rate or elastic modulus change coefficient corresponding to the change in air bladder thickness during performance degradation testing. The oxidation rate reflects the rate of aging of the air bladder material over time, while the elastic modulus change coefficient reflects the change in the elasticity of the air bladder material. In this invention, by combining the physical parameters of the air bladder and the stress relaxation data, the degradation data of the air bladder can be calculated, including the degradation rate and degree of degradation, for evaluating the durability and service life of the air bladder.

[0078] Finally, the data on the change in gas pressure inside the air spring and the data on the degradation of the air bladder are used as the physical parameters of the spring's stiffness for subsequent calculations of dynamic stiffness performance degradation data.

[0079] Step 102: Input the physical parameters of the spring stiffness into the air spring performance degradation model to obtain the dynamic stiffness performance degradation data of the air spring to be tested;

[0080] The dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data; the air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model. The dynamic stiffness frequency correlation model is used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters; the dynamic stiffness amplitude correlation model is used to calculate the dynamic stiffness amplitude correlation data based on the air bladder degradation data in the spring stiffness physical parameters.

[0081] In this invention, the physical parameters of spring stiffness obtained through the performance degradation test process are input into the air spring performance degradation model. These parameters include data on changes in internal gas pressure and airbag degradation. Based on the air spring performance degradation model, the performance degradation of the air spring under different conditions can be predicted according to the physical parameters of spring stiffness.

[0082] In this invention, the air spring performance degradation model comprises two main parts: a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model. The dynamic stiffness frequency correlation model calculates the change in dynamic stiffness of the air spring at different frequencies. Its input data mainly consists of the change in internal gas pressure within the air spring, a key physical parameter for spring stiffness. The output data is the dynamic stiffness frequency correlation data, reflecting the dynamic stiffness performance of the air spring under different frequency excitations. Based on the input gas pressure change data, combined with the physical characteristics and dynamic response characteristics of the air spring, the dynamic stiffness frequency correlation model calculates the dynamic stiffness values ​​at different frequencies.

[0083] The dynamic stiffness amplitude correlation model is used to calculate the change in dynamic stiffness of an air spring under different amplitudes. Its input data mainly consists of air bladder degradation data from the spring stiffness physical parameters, and its output data is dynamic stiffness amplitude correlation data, reflecting the dynamic stiffness performance of the air spring under different amplitude excitations. The dynamic stiffness amplitude correlation model calculates the dynamic stiffness values ​​at different amplitudes based on the input air bladder degradation data, taking into account factors such as air bladder material aging and changes in elastic modulus.

[0084] In this invention, the calculated stiffness-frequency correlation data and dynamic stiffness-amplitude correlation data can be used for performance degradation analysis and characterization of air springs. For example, analyzing the dynamic stiffness variation trend of an air spring at different frequencies, if the dynamic stiffness decreases significantly with increasing frequency, it may indicate performance degradation of the air spring under high-frequency excitation. Alternatively, analyzing the dynamic stiffness-amplitude correlation data and observing the dynamic stiffness variation of the air spring at different amplitudes, if the dynamic stiffness decreases significantly with increasing amplitude, it may mean performance degradation of the air spring under large-amplitude excitation.

[0085] The air spring performance degradation characterization method provided by this invention combines the dynamic changes in the internal gas pressure of the air spring, the nonlinear transmission characteristics of the rubber airbag, and the air spring performance degradation phenomenon caused by multiple internal and external factors. It provides more accurate air spring performance prediction and degradation analysis results. It is not only applicable to the dynamic analysis of air springs, but also provides theoretical basis and technical support for the design and control optimization of vehicle suspension systems.

[0086] Based on the above embodiments, the airbag stress relaxation data of the airbag is obtained through the following steps:

[0087] Based on the Arrhenius model, the coefficient of change of elastic modulus is obtained according to the thermal acceleration coefficient and the thickness of the airbag during the performance degradation test.

[0088] Based on the airbag thickness change status information and diffusion coefficient, the airbag oxidation rate corresponding to the airbag thickness when subjected to the force is obtained.

[0089] Based on the applied force and the elastic modulus variation coefficient, the airbag stress relaxation data corresponding to the airbag thickness when subjected to the applied force is calculated; or, based on the initial airbag stress data and initial oxidation rate corresponding to the airbag in the initial state, and the airbag oxidation rate, the airbag stress relaxation data corresponding to the airbag thickness when subjected to the applied force is calculated.

[0090] The thermal acceleration coefficient and the diffusion coefficient are obtained based on a few-shot deep learning network model trained by a deep learning network, according to the input preset air spring test conditions. The few-shot deep learning network model is trained using sample dynamic stiffness performance degradation data generated under sample air spring test conditions, as well as the thermal acceleration coefficient label and diffusion coefficient label corresponding to the sample dynamic stiffness performance degradation data.

[0091] During long-term use, the stiffness of air springs gradually weakens due to factors such as rubber material aging, cord breakage, and air leakage. This invention proposes a rubber aging model based on diffusion-limited oxidation, which can accurately describe the performance degradation process of rubber materials over time and with changes in ambient temperature. This rubber aging model, through fitting experimental data, yields curves showing the rubber oxidation rate and mechanical property changes under different environmental conditions. Besides aging, air springs exhibit stress relaxation under fatigue loads, causing the relationship between material deformation and the force it bears to become non-linear. This invention establishes a degradation model of modulus with temperature using stress relaxation theory and the Arrhenius model.

[0092] In this invention, the performance of rubber airbags and other structural components of air springs may degrade after prolonged operation due to internal factors such as loading history, structural design, and sealing; it may also be affected by external factors such as loading history, environment, and actual operating conditions. Therefore, understanding the underlying mechanisms and characterizing this degradation presents significant challenges, which is a key issue this invention aims to address. Similar to the aforementioned points, this invention strives to improve the generalizability and interpretability of the proposed theory by incorporating physically meaningful model derivations. Various models exist for accelerating rubber aging, and current research generally considers the effects of oxygen to be the cause. Therefore, this invention employs a diffusion-limited oxidation model, which better aligns with physical laws, to characterize the performance degradation of rubber and to calculate the airbag stress relaxation data within the air spring.

[0093] Based on the above embodiments, the airbag stress relaxation data is calculated using the following formula:

[0094] ;

[0095] The formula for calculating the oxidation rate of the airbag is:

[0096] ;

[0097] The Arrhenius model, which describes the change of modulus with temperature, is used to calculate the coefficient of change of elastic modulus. The formula for calculating the coefficient of change of elastic modulus is as follows:

[0098] ;

[0099] in, This indicates that the airbag has a thickness of [missing information]. The corresponding airbag stress relaxation data at that time, This indicates that the airbag has a thickness of [missing information]. The magnitude of the force applied at that time, This indicates the preset air spring test amplitude. This represents the coefficient of variation of the elastic modulus. This represents the initial airbag stress data. This represents the initial oxidation rate; This indicates the thickness displacement of the air spring under test before it is subjected to the applied force. This indicates the thickness displacement of the air spring under test after being subjected to the applied force; This represents the diffusion coefficient. Represents the diffusion function; The integral factor represents a diffusion-related differential parameter or a dimensionless parameter related to the distribution of gas molecules. In the integral formula, It is an integral variable, ranging from [0, 1], and is related to the calculation of the diffusion rate; The gas compressibility factor is used to correct for the behavior of gases under non-ideal conditions such as high pressure or high temperature. It describes the degree to which a gas deviates from its ideal state and is usually derived experimentally or from equations of state (such as the Van der Waals equation). In this invention, The behavior of the gas during expansion has been adjusted to make the calculations more consistent with actual working conditions; It represents the acceleration coefficient, which is used to describe the correlation between the performance degradation of an air spring and external conditions (such as vibration, frequency, etc.), and reflects the nonlinear acceleration effect; This indicates the amount of change in airbag thickness, or the corresponding geometric parameters after airbag deformation; The thermal acceleration coefficient describes the effect of temperature change on the change of the elastic modulus of the air spring, reflecting the role of thermodynamic factors in performance degradation; The thermal influence factor can be understood as a parameter related to the thermal effect of air spring materials. It is used to describe the effect of temperature on the change of elastic modulus and the performance degradation of air springs. Its mechanism of action is as follows: an increase in temperature may cause material softening, airbag volume expansion, and air pressure changes, which in turn affect the elastic modulus of the air spring. H, as an empirical coefficient or a parameter related to the thermal properties of materials, can correct the accelerating effect of temperature changes on performance degradation. This indicates the equivalent activation energy of the airbag. Represents the gas constant. This indicates absolute temperature.

[0100] Based on the above embodiments, the method further includes:

[0101] Based on the residual results output by the few-sample deep learning network model, the dynamic stiffness performance degradation data is corrected to obtain the corrected dynamic stiffness performance degradation data.

[0102] In this invention, after clarifying the variable-parameter dynamic characteristics (such as pressure, volume, etc.) of the air spring and the material-phenomenon mapping relationship, it is necessary to clarify the mechanism of unknown parameter changes under different environments. For example, the diffusion coefficient D and thermal acceleration coefficient C2 in the stress relaxation data of the airbag mentioned above are physical parameters that need to be determined. Because air springs are affected by environmental, material, and loading history factors, it is difficult to directly derive the general law of parameter changes from forward physics modeling, and it also faces practical problems such as the difficulty of obtaining actual test samples and the long testing cycle. Therefore, this invention constructs a small-sample deep learning network. Figure 4 This is a schematic diagram illustrating the construction process of the few-shot deep learning network provided by the present invention, which can be referred to. Figure 4As shown, the thermal acceleration coefficient and diffusion coefficient described in this invention are not obtained directly through experimental measurement, but are predicted by a small sample deep learning network model.

[0103] During training, firstly, a set of dynamic stiffness performance degradation data collected under specific sample air spring test conditions is required. This data records the specific values ​​of the dynamic stiffness degradation of the air spring over time under different temperature, pressure, and time conditions. In addition to the aforementioned dynamic stiffness performance degradation data, the true or accurate measured values ​​of the thermal acceleration coefficient and diffusion coefficient corresponding to each set of data are also needed as labels for the model. These label data are obtained through specialized experimental measurements or known theoretical calculations.

[0104] Furthermore, using the aforementioned sample data and corresponding label data, a deep learning network model is constructed. Since the model is designed for small sample sizes, special attention is paid to generalization ability and preventing overfitting, employing techniques such as data augmentation, regularization, and transfer learning. Through iterative training, the model gradually learns to predict the corresponding thermal acceleration coefficient and diffusion coefficient from the input preset air spring test conditions.

[0105] Once the model is trained and its predictive accuracy is verified, it can be used in practical applications to quickly and accurately predict the required thermal acceleration coefficient and diffusion coefficient based on given air spring test conditions. This greatly simplifies the parameter acquisition process and improves research efficiency and accuracy. This invention trains the model with a small amount of experimental data, automatically identifying performance degradation parameters of air springs at different fatigue stages, improving the model's adaptability and accuracy, and reducing reliance on large amounts of experimental data.

[0106] Preferably, in this invention, the physical model is also calculated and corrected based on the residuals of a small sample measured dataset, which can be referred to... Figure 4 As shown, by conducting a power indicator test on the air spring and combining the model residuals to perform reverse identification on the above parameters, the corrected dynamic stiffness performance degradation data are obtained. For example, the corrected dynamic stiffness frequency correlation of the air spring can be expressed as:

[0107] ;

[0108] in, This represents the dynamic stiffness frequency correlation data before correction. This represents the corresponding residual result. The explanations for the other parameters in this formula will be provided in subsequent steps.

[0109] Based on the above embodiments, the formula for the dynamic stiffness frequency correlation model is as follows:

[0110] ;

[0111] ;

[0112] in, Indicates angular frequency as Dynamic stiffness frequency correlation data; This indicates the stiffness generated by the high-pressure gas inside the air spring under test; This refers to the equivalent damping generated by the heat exchange between the internal gas of the airbag and the gas inside the auxiliary air chamber and the outside environment. This represents the equivalent stiffness generated by the additional air chamber; This represents the equivalent damping generated by the high-pressure gas flowing through the connecting hole inside the air spring under test; Indicates specific heat ratio. This indicates the initial pressure of the gas inside the air spring being tested. This indicates the effective area of ​​the gas inside the air spring. This indicates the initial volume of the gas inside the air spring being tested. Indicates constant-volume heat capacity. This indicates the mass of gas inside the air spring being tested; It represents the equivalent heat transfer coefficient, which is the coefficient of heat exchange between the internal gas and the external environment; This indicates that the air spring under test is subjected to the force described above, and the thickness of the air bladder is... The volume of the internal gas at that time; Represents the ideal gas constant. This represents the correction factor for the equivalent area of ​​the throttling orifice. This indicates the initial temperature of the gas inside the air spring being tested. This represents the equivalent damping coefficient of the throttling orifice.

[0113] Figure 5 This is a schematic diagram of the air spring performance degradation characterization method provided by the present invention. For an interpretable dynamic hybrid characterization model of air spring performance degradation under small sample and multiple internal and external factors, the technical route proposed in this invention can be referred to... Figure 5 As shown. In this invention, firstly, for the air spring mechanism model, its dynamic stiffness expression is constructed:

[0114] ;

[0115] in, For the thermodynamic model of an air spring, Indicates gas frequency correlation. Indicates effective area stiffness; The dynamic stiffness of an air spring can be divided into two main parts: frequency dependence, which is related to the internal gas pressure; and stiffness, which is related to the effective area; and amplitude dependence, which is related to the material and structural properties of the rubber air spring. The forward derivation of the dynamic stiffness expression, namely the formulas for the dynamic stiffness frequency dependence model and the dynamic stiffness amplitude dependence model, is based on thermodynamic and viscoelastic theories.

[0116] In this invention, the effective area of ​​the gas inside the air spring can be determined by considering the rubber airbag as a multi-layer composite laminate, analyzing its mechanical properties in conjunction with material design and parameters, and then calculating the stiffness. The formula for calculating the effective area of ​​the gas inside the air spring is as follows:

[0117] ;

[0118] in, A parameter representing the effect of the air spring's height on its effective area. Indicates the height of the air spring. The parameters representing the influence of the initial working height of the air spring are: This represents the corresponding element in the reduced stiffness matrix of the pivot shaft; The first pressure change effect parameter is determined by the airbag radius after the pressure change and the initial radius when the pressure inside the airbag is small and no longer undergoes significant radial deformation with pressure change. The second parameter representing the effect of pressure change is determined by the airbag radius and shape parameters after the pressure change. It represents relative pressure.

[0119] The dynamic stiffness amplitude correlation of an air spring is mainly provided by the strain amplitude characteristics of the rubber material. This invention, referencing the material-level Kraus model, studies the macroscopic amplitude correlation of air springs. Combining the principle of minimum complementary energy, the formula for the constructed dynamic stiffness amplitude correlation model is as follows:

[0120] ;

[0121] in, Indicates the test amplitude is Dynamic stiffness amplitude correlation data at time; Indicates the effective radius of the air spring. This represents the correction coefficient for the correlation between dynamic stiffness and amplitude. This indicates the thickness of the reinforcing composite material inside the air spring under test. This indicates the volume fraction of the cord in the air bladder of the air spring under test; This indicates the Young's modulus or damage modulus of the cord in the air spring under test. This represents the airbag stress relaxation data; Indicates the inclination angle of the cord. This indicates the characteristic amplitude of the airbag. This indicates the type of rubber material used in the airbag. This indicates the total thickness of the air bladder in the air spring under test. This indicates the deflection radius of the air spring under test. This indicates the piston tilt angle.

[0122] In this invention, one of the main sources of the dynamic stiffness of the air spring is the reaction force of the internal high-pressure gas to the external vibration response. In the stiffness amplitude correlation model, the heat exchange coefficient between the gas and the external environment, as well as the frequency correlation, are derived using thermodynamic principles. The expression for dynamic stiffness is based on a combination of isothermal and adiabatic processes, considering the effects of static and dynamic pressure changes on the air spring stiffness.

[0123] In the formula of the dynamic stiffness amplitude correlation model, all parameters can be directly measured, ensuring the physical interpretability and generalizability of the model. For example, the total thickness of the airbag is... t r The thickness of the internal reinforced composite material layer is t s Piston tilt angle φ s The radius of the drooping ear is r E f and E m The Young's modulus or loss modulus, representing the cord and rubber materials respectively, can be obtained using the calculation formulas for the stress relaxation data of the airbag mentioned above. Furthermore, its nonlinear transmission characteristics can be determined using the dynamic stiffness amplitude correlation model.

[0124] In this invention, another major source of stiffness in the air spring is the nonlinear stress-strain response of the rubber air bladder. This invention employs the Kraus model to study the characteristics of the rubber strain amplitude and, combined with the mechanical properties of the cord composite structure, derives the amplitude correlation of dynamic stiffness, thus solving the nonlinear problem of the dynamic stiffness of the air spring changing with amplitude. Compared to traditional linear models, the model of this invention can more accurately characterize the dynamic performance of the air spring under different amplitudes and frequencies, comprehensively considering the influence of multiple internal and external factors (such as temperature, humidity, aging, stress relaxation, etc.) on the stiffness of the air spring structure design, including cord tilt angle and Young's modulus of the material.

[0125] This invention provides an interpretable dynamic hybrid characterization method for air spring performance degradation, addressing the problem of insufficient dynamic characteristic characterization in existing air spring performance degradation analyses. This invention combines the dynamic changes in internal gas pressure of the air spring, the nonlinear transmission characteristics of the rubber air bladder, and performance degradation phenomena caused by multiple internal and external factors, proposing a hybrid characterization method driven by both mechanistic data.

[0126] This invention, through a combination of physical modeling and deep learning, first constructs a dynamic stiffness model for an air spring. This model comprehensively considers factors such as frequency dependence, amplitude dependence, and heat exchange. Specifically, based on thermodynamics and viscoelasticity theories, the dynamic stiffness of the air spring consists of two parts: one part is related to the high-pressure gas pressure and heat exchange, and the other part is related to the change in the effective area of ​​the rubber air bladder. This invention clarifies the influence of amplitude on dynamic stiffness by modeling the multilayer composite material structure of the rubber air bladder and derives the nonlinear strain characteristics of the rubber material using the principle of minimum complementary energy.

[0127] Based on the established physical model, this invention introduces few-sample deep learning technology to identify and correct unknown parameters in the performance degradation process of air springs. Using fatigue test data of air springs under actual operating conditions, this invention uses a deep learning network to train the model, automatically identifying parameter changes under different operating conditions, thereby compensating and optimizing the physical model. Ultimately, through a dual-driven model of mechanism and data, the dynamic characteristics and performance degradation process of the air spring are accurately characterized. This invention, in a small-sample data environment combined with actual operating conditions, provides more accurate air spring performance prediction and degradation analysis results through a combination of deep learning and physical models. It is not only applicable to the dynamic analysis of air springs but also provides theoretical basis and technical support for the design and control optimization of vehicle suspension systems. This invention proposes a method combining few-sample deep learning, training the model with a small amount of experimental data to identify the parameter variation patterns of air springs under different operating conditions. The system adjusts the modeling parameters in real time based on real-time monitoring data, ensuring the dynamic updating and correction of the physical model.

[0128] Figure 6 This is a schematic diagram showing the comparison between the performance degradation test results and theoretical values ​​of the air spring provided by the present invention, which can be used as a reference. Figure 6As shown, after clarifying the variable parameter mechanism and dual-drive model, the dynamic characteristics of the air suspension system can be studied, and highly reliable collaborative compensation control can be implemented for the variable parameter air suspension system under random time-varying external loads. In the experiment, the same air spring was used for dynamic stiffness comparison after initial state, 1.8 million, 3.6 million, and 5.4 million fatigue cycles. The experiment used a 1Hz excitation frequency, and the amplitude of the air spring was 1mm, 2mm, 5mm, 10mm, and 20mm, respectively. In this invention, the data-driven part acquires the experimental data of the air spring at different fatigue stages (such as 1.8 million, 3.6 million, and 4.5 million cycles), and uses deep learning technology to analyze parameters that are difficult to measure, such as the heat exchange coefficient inside the air spring and the changes in the mechanical properties of the rubber material, thereby compensating for the errors caused by unknown parameters in the physical model.

[0129] This invention enables accurate characterization of air spring performance degradation in a small sample data environment and can be used to predict the remaining life of air springs. By monitoring the real-time operating status of air springs, it can automatically identify the operating conditions of air springs and adjust the suspension control strategy of air springs according to changes in operating conditions, so as to ensure the safety and comfort of vehicles under different loads and road conditions.

[0130] The air spring performance degradation characterization system provided by the present invention will be described below. The air spring performance degradation characterization system described below can be referred to in correspondence with the air spring performance degradation characterization method described above.

[0131] Figure 7 A schematic diagram of the air spring performance degradation characterization system provided by the present invention is shown below. Figure 7 As shown, this invention provides an air spring performance degradation characterization system, including a test parameter acquisition module 701 and a performance degradation characterization module 702. The test parameter acquisition module 701 is used to acquire the spring stiffness physical parameters generated during the performance degradation test of the air spring under test based on an air suspension test system. The performance degradation characterization module 702 is used to input the spring stiffness physical parameters into an air spring performance degradation model to acquire the dynamic stiffness performance degradation data of the air spring under test. The dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data. The air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model. The dynamic stiffness frequency correlation model is used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters. The dynamic stiffness amplitude correlation model is used to calculate the dynamic stiffness amplitude correlation data based on the airbag degradation data in the spring stiffness physical parameters.

[0132] The air spring performance degradation characterization system provided by this invention combines the dynamic changes in the internal gas pressure of the air spring, the nonlinear transmission characteristics of the rubber airbag, and the air spring performance degradation phenomenon caused by multiple internal and external factors, to provide more accurate air spring performance prediction and degradation analysis results. It is not only applicable to the dynamic analysis of air springs, but also provides theoretical basis and technical support for the design and control optimization of vehicle suspension systems.

[0133] The system provided in this embodiment of the invention is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0134] Figure 8 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 8 As shown, the electronic device may include: a processor 801, a communications interface 802, a memory 803, and a communication bus 804, wherein the processor 801, the communications interface 802, and the memory 803 communicate with each other through the communication bus 804. The processor 801 can call logic instructions in the memory 803 to execute an air spring performance degradation characterization method. This method includes: acquiring the spring stiffness physical parameters generated during the performance degradation test of the air spring under test based on an air suspension testing system; inputting the spring stiffness physical parameters into an air spring performance degradation model to obtain the dynamic stiffness performance degradation data of the air spring under test; wherein the dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data; the air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model, wherein the dynamic stiffness frequency correlation model is used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters; and the dynamic stiffness amplitude correlation model is used to calculate the dynamic stiffness amplitude correlation data based on the airbag degradation data in the spring stiffness physical parameters.

[0135] Furthermore, the logical instructions in the aforementioned memory 803 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0136] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein when the program instructions are executed by a computer, the computer is able to execute the air spring performance degradation characterization method provided by the above methods, the method comprising: acquiring the spring stiffness physical parameters generated by the air spring under test during the performance degradation test based on an air suspension testing system; inputting the spring stiffness physical parameters into an air spring performance degradation model to acquire the dynamic stiffness performance degradation data of the air spring under test; wherein the dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data; the air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model, the dynamic stiffness frequency correlation model being used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters; the dynamic stiffness amplitude correlation model being used to calculate the dynamic stiffness amplitude correlation data based on the airbag degradation data in the spring stiffness physical parameters.

[0137] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the air spring performance degradation characterization method provided in the above embodiments. The method includes: acquiring spring stiffness physical parameters generated during the performance degradation test of the air spring under test using an air suspension testing system; inputting the spring stiffness physical parameters into an air spring performance degradation model to obtain dynamic stiffness performance degradation data of the air spring under test; wherein the dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data; the air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model, wherein the dynamic stiffness frequency correlation model is used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters; and the dynamic stiffness amplitude correlation model is used to calculate the dynamic stiffness amplitude correlation data based on the airbag degradation data in the spring stiffness physical parameters.

[0138] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0139] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for characterizing the performance degradation of an air spring, characterized in that, include: Based on the air suspension testing system, the physical parameters of spring stiffness generated during the performance degradation test of the air spring under test are obtained. The physical parameters of spring stiffness are input into the air spring performance degradation model to obtain the dynamic stiffness performance degradation data of the air spring under test. The dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data; the air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model. The dynamic stiffness frequency correlation model is used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters; the dynamic stiffness amplitude correlation model is used to calculate the dynamic stiffness amplitude correlation data based on the air bladder degradation data in the spring stiffness physical parameters. The formula for the dynamic stiffness frequency correlation model is as follows: ； ； in, Indicates angular frequency as Dynamic stiffness frequency correlation data; This indicates the stiffness generated by the high-pressure gas inside the air spring under test; This refers to the equivalent damping generated by the heat exchange between the internal gas of the airbag and the gas inside the auxiliary air chamber and the outside environment. This represents the equivalent stiffness generated by the additional air chamber; This represents the equivalent damping generated by the high-pressure gas flowing through the connecting hole inside the air spring under test; Indicates specific heat ratio. This indicates the initial pressure of the gas inside the air spring being tested. This indicates the effective area of ​​the gas inside the air spring. This indicates the initial volume of the gas inside the air spring being tested. Indicates constant-volume heat capacity. This indicates the mass of gas inside the air spring being tested. Indicates the equivalent heat transfer coefficient; This indicates that the air spring under test is subjected to a force, and the thickness of the airbag is... The volume of the internal gas at that time; Represents the ideal gas constant. This represents the correction factor for the equivalent area of ​​the throttling orifice. This indicates the initial temperature of the gas inside the air spring being tested. Indicates the equivalent damping coefficient of the throttling orifice; The specific formula for the dynamic stiffness amplitude correlation model is as follows: ； in, Indicates the test amplitude is Dynamic stiffness amplitude correlation data at time; Indicates the effective radius of the air spring. This represents the correction coefficient for the correlation between dynamic stiffness and amplitude. This indicates the thickness of the reinforcing composite material inside the air spring under test. This indicates the volume fraction of the cord in the air bladder of the air spring under test; This indicates the Young's modulus or damage modulus of the cord in the air spring under test. This represents airbag stress relaxation data; Indicates the inclination angle of the cord. This indicates the characteristic amplitude of the airbag. This indicates the type of rubber material used in the airbag. This indicates the total thickness of the air bladder in the air spring under test. This indicates the deflection radius of the air spring under test. This indicates the piston tilt angle.

2. The method for characterizing the performance degradation of an air spring according to claim 1, characterized in that, The air suspension testing system includes a hydraulic excitation device, a force sensor, and a pressure sensor. The hydraulic excitation device is positioned directly below the air spring under test and is used to apply a periodic force to the air spring according to preset air spring test conditions to control the compression and extension states of the air spring. The preset air spring test conditions include a preset hydraulic excitation frequency, a preset number of fatigue test cycles, and a preset air spring test amplitude. The force sensor is positioned directly above the air spring under test and is used to measure the changes in the air spring under test when subjected to force. The air suspension testing system acquires the physical parameters of the spring stiffness generated during the performance degradation test of the air spring under test, including: When the air spring under test is subjected to the periodic force applied by the hydraulic excitation device, the internal gas pressure data, internal gas temperature data, and effective area of ​​the internal gas of the air spring are collected as the internal gas of the air spring changes with the force. Based on the internal gas pressure data, internal gas temperature data, and effective area of ​​the internal gas in the air spring, the internal gas pressure change data of the air spring is obtained. Obtain the physical parameters of the air bladder in the air spring to be tested; When the air spring under test is subjected to the periodic force applied by the hydraulic excitation device, the airbag stress relaxation data of the airbag is obtained, wherein the airbag stress relaxation data is calculated based on the airbag oxidation rate or elastic modulus change coefficient corresponding to the airbag thickness during the performance degradation test. Based on the physical parameters of the airbag and the stress relaxation data of the airbag, the degradation data of the airbag is obtained. The physical parameters of the spring stiffness are obtained based on the data of gas pressure change inside the air spring and the data of air bladder degradation.

3. The method for characterizing the performance degradation of an air spring according to claim 2, characterized in that, The airbag stress relaxation data is obtained through the following steps: Based on the Arrhenius model, the coefficient of change of elastic modulus is obtained according to the thermal acceleration coefficient and the thickness of the airbag during the performance degradation test. Based on the airbag thickness change status information and diffusion coefficient, the airbag oxidation rate corresponding to the airbag thickness when subjected to the force is obtained. Based on the applied force and the elastic modulus variation coefficient, the airbag stress relaxation data corresponding to the airbag thickness when subjected to the applied force is calculated; or, based on the initial airbag stress data and initial oxidation rate corresponding to the airbag in the initial state, and the airbag oxidation rate, the airbag stress relaxation data corresponding to the airbag thickness when subjected to the applied force is calculated. The thermal acceleration coefficient and the diffusion coefficient are obtained based on a few-shot deep learning network model trained by a deep learning network, according to the input preset air spring test conditions. The few-shot deep learning network model is trained using sample dynamic stiffness performance degradation data generated under sample air spring test conditions, as well as the thermal acceleration coefficient label and diffusion coefficient label corresponding to the sample dynamic stiffness performance degradation data.

4. The method for characterizing the performance degradation of an air spring according to claim 3, characterized in that, The method further includes: Based on the residual results output by the few-sample deep learning network model, the dynamic stiffness performance degradation data is corrected to obtain the corrected dynamic stiffness performance degradation data.

5. The method for characterizing the performance degradation of an air spring according to claim 3, characterized in that, The airbag stress relaxation data is calculated using the following formula: ； The formula for calculating the oxidation rate of the airbag is: ； The formula for calculating the coefficient of change of the elastic modulus is: ； in, This indicates that the airbag has a thickness of [missing information]. The corresponding airbag stress relaxation data at that time This indicates that the airbag has a thickness of [missing information]. The magnitude of the force applied at that time, This indicates the preset air spring test amplitude. This represents the coefficient of variation of the elastic modulus. This represents the initial airbag stress data. This represents the initial oxidation rate; This indicates the thickness displacement of the air spring under test before it is subjected to the applied force. This indicates the thickness displacement of the air spring under test after being subjected to the applied force; This represents the diffusion coefficient. Represents the diffusion function. Indicates the integrating factor. Indicates the gas compressibility factor. Indicates the acceleration coefficient. This indicates the change in airbag thickness. This represents the thermal acceleration coefficient. Indicates the thermal influence factor. This indicates the equivalent activation energy of the airbag. Represents the gas constant. This indicates absolute temperature.

6. A system for characterizing the performance degradation of an air spring, characterized in that, include: The test parameter acquisition module is used to acquire the physical parameters of spring stiffness generated by the air spring under test during the performance degradation test based on the air suspension test system. The performance degradation characterization module is used to input the physical parameters of the spring stiffness into the air spring performance degradation model to obtain the dynamic stiffness performance degradation data of the air spring under test. The dynamic stiffness performance degradation data includes dynamic stiffness frequency correlation data and dynamic stiffness amplitude correlation data; the air spring performance degradation model includes a dynamic stiffness frequency correlation model and a dynamic stiffness amplitude correlation model. The dynamic stiffness frequency correlation model is used to calculate the stiffness frequency correlation data based on the air spring internal gas pressure change data in the spring stiffness physical parameters; the dynamic stiffness amplitude correlation model is used to calculate the dynamic stiffness amplitude correlation data based on the air bladder degradation data in the spring stiffness physical parameters. The formula for the dynamic stiffness frequency correlation model is as follows: ； ； in, Indicates angular frequency as Dynamic stiffness frequency correlation data; This indicates the stiffness generated by the high-pressure gas inside the air spring under test; This refers to the equivalent damping generated by the heat exchange between the internal gas of the airbag and the gas inside the auxiliary air chamber and the outside environment. This represents the equivalent stiffness generated by the additional air chamber; This represents the equivalent damping generated by the high-pressure gas flowing through the connecting hole inside the air spring under test; Indicates specific heat ratio. This indicates the initial pressure of the gas inside the air spring being tested. This indicates the effective area of ​​the gas inside the air spring. This indicates the initial volume of the gas inside the air spring being tested. Indicates constant-volume heat capacity. This indicates the mass of gas inside the air spring being tested. Indicates the equivalent heat transfer coefficient; This indicates that the air spring under test is subjected to a force, and the thickness of the airbag is... The volume of the internal gas at that time; Represents the ideal gas constant. This represents the correction factor for the equivalent area of ​​the throttling orifice. This indicates the initial temperature of the gas inside the air spring being tested. Indicates the equivalent damping coefficient of the throttling orifice; The specific formula for the dynamic stiffness amplitude correlation model is as follows: ； in, Indicates the test amplitude is Dynamic stiffness amplitude correlation data at time; Indicates the effective radius of the air spring. This represents the correction coefficient for the correlation between dynamic stiffness and amplitude. This indicates the thickness of the reinforcing composite material inside the air spring under test. This indicates the volume fraction of the cord in the air bladder of the air spring under test; This indicates the Young's modulus or damage modulus of the cord in the air spring under test. This represents airbag stress relaxation data; Indicates the inclination angle of the cord. This indicates the characteristic amplitude of the airbag. This indicates the type of rubber material used in the airbag. This indicates the total thickness of the air bladder in the air spring under test. This indicates the deflection radius of the air spring under test. This indicates the piston tilt angle.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the air spring performance degradation characterization method as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the air spring performance degradation characterization method as described in any one of claims 1 to 5.

Images