Spring fatigue life test system and dynamic balance evaluation method

Abstract

The application relates to the technical field of spring measurement, and discloses a spring fatigue life test system and a dynamic balance evaluation method. The system comprises a reference length acquisition module, a dynamic balance state determination module, a working condition spectrum loading module, a stress-strain response data acquisition module, a damage feature extraction module and a life evaluation module. The reference length in a free state is acquired. A step-by-step increasing preload is applied. When the deformation rate of the to-be-tested spring is detected to be smaller than a set threshold value, it is determined that the to-be-tested spring reaches a dynamic balance state. In the dynamic balance state, the to-be-tested spring is subjected to load loading based on a variable-amplitude load time sequence signal, and stress-strain response data is synchronously collected. Stress features of the to-be-tested spring in a load cycle are identified. A life evaluation report is generated based on the stress features. The application can improve the accuracy of spring fatigue life.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a spring fatigue life testing system and a dynamic balance evaluation method. Background Technology

[0002] In the field of spring fatigue life testing, existing technologies often struggle to accurately simulate the complex load environments encountered during actual service, leading to significant deviations between test conditions and real-world operating conditions. This results in fatigue life data that fails to accurately reflect the spring's durability in actual use. Furthermore, traditional testing methods often fail to accurately determine whether the spring has reached a stable dynamic equilibrium state during preloading, impacting the accuracy of subsequent stress-strain response data acquisition and ultimately reducing the reliability of fatigue life assessment results.

[0003] Furthermore, existing technologies lack sufficient detail in analyzing hysteresis loops when extracting spring damage characteristics, making it difficult to accurately identify key parameters such as peak stress, trough stress, and the area enclosed by the hysteresis loop. This results in significant errors in fitting the damage accumulation trajectory. Moreover, the lack of an effective load compensation mechanism during load application leads to a large deviation between the actual applied load and the target load, further exacerbating the inaccuracy of fatigue life assessment and failing to meet the requirements of high-precision testing. Summary of the Invention

[0004] This invention provides a spring fatigue life testing system and a dynamic balance evaluation method to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the spring fatigue life testing system and dynamic equilibrium evaluation method provided by the present invention are characterized by comprising a reference length acquisition module, a dynamic equilibrium state determination module, a working condition spectrum loading module, a stress-strain response data acquisition module, a damage feature extraction module, and a life evaluation module, wherein:

[0006] The reference length acquisition module is used to acquire the reference length of the spring to be tested in a free state;

[0007] The dynamic equilibrium state determination module is used to apply a stepped increasing preload to the spring under test. When the deformation rate of the spring under test is detected to be less than a set threshold, it is determined that the spring under test has reached a dynamic equilibrium state.

[0008] The working condition spectrum loading module is used to generate the variable amplitude load time sequence signal of the spring under test based on the vibration spectrum characteristics of the spring under test in the service environment.

[0009] The stress-strain response data acquisition module is used to load the spring under test based on the variable amplitude load time sequence signal under the dynamic equilibrium state, and simultaneously acquire the stress-strain response data of the spring under test.

[0010] The damage feature extraction module is used to identify the peak stress, valley stress and hysteresis loop area of ​​the spring under test during the load cycle based on the hysteresis loop inflection point in the stress-strain response data.

[0011] The life assessment module is used to generate a life assessment report for the spring under test based on the peak stress, the valley stress, and the area enclosed by the loop.

[0012] In a preferred embodiment, when the reference length acquisition module acquires the reference length of the spring to be measured in its free state, it is specifically used for:

[0013] The unloaded spring under test is mounted on the base of the test platform;

[0014] The reference length data of the spring under test is generated based on the projected distance between the spring under test and the reference surface in the base of the test platform.

[0015] In a preferred embodiment, when the dynamic equilibrium state determination module applies a stepped increasing preload to the spring under test and determines that the spring under test has reached a dynamic equilibrium state when the deformation rate of the spring under test is less than a set threshold, it is specifically used for:

[0016] The incremental step size of the stepped load is determined based on the rated load parameters of the spring under test;

[0017] The preload is applied to the spring under test step by step according to the incremental step size;

[0018] During the load holding period, the axial deformation data of the spring under test is collected in real time;

[0019] The real-time deformation rate curve of the spring under test is generated based on the axial deformation data;

[0020] When the real-time deformation rate curve remains below the rate threshold, the spring under test is determined to have reached a dynamic equilibrium state.

[0021] In a preferred embodiment, when the operating condition spectrum loading module generates the variable amplitude load time sequence signal of the spring under test based on the vibration spectrum characteristics of the spring under test in the service environment, it is specifically used for:

[0022] Collect vibration acceleration data of the spring under test under typical working conditions;

[0023] The vibration acceleration data is converted into the frequency domain to obtain the main energy distribution frequency band of the spring under test;

[0024] The load amplitude variation range of the spring under test is determined based on the main energy distribution frequency band.

[0025] A random amplitude modulation signal is generated based on the range of load amplitude variation, and a variable amplitude load timing signal of the spring under test is constructed based on the random amplitude modulation signal.

[0026] In a preferred embodiment, when the load spectrum loading module generates a random amplitude modulation signal based on the load amplitude variation range and constructs the variable amplitude load timing signal of the spring under test based on the random amplitude modulation signal, it is specifically used for:

[0027] Establish a mapping relationship between the range of load amplitude variation and the time series;

[0028] A basic sinusoidal carrier signal is constructed under the constraints of the aforementioned mapping relationship;

[0029] The amplitude of the basic sinusoidal carrier signal is randomized.

[0030] The randomized base sinusoidal carrier signal is subjected to time-domain smoothing filtering to obtain the variable amplitude load timing signal of the spring under test.

[0031] In a preferred embodiment, when the stress-strain response data acquisition module performs load loading on the spring under test based on the amplitude load timing signal under the dynamic equilibrium state, and simultaneously acquires the stress-strain response data of the spring under test, it is specifically used for:

[0032] The spring under test is loaded according to the variable amplitude load timing signal;

[0033] Based on the deviation between the target load value corresponding to the variable amplitude load time sequence signal and the actual load value during the load loading, the compensation value of the load is calculated, wherein the calculation formula of the compensation value is as follows:

[0034]

[0035] In the formula, The compensation value is... This is the proportional gain coefficient. The deviation amount, This is the integral gain coefficient. The time factor;

[0036] The compensation value is superimposed on the driving current command of the load, and the stress-strain response data of the spring under test is collected simultaneously.

[0037] In a preferred embodiment, when the damage feature extraction module identifies the peak stress, valley stress, and hysteresis loop enclosed area of ​​the spring under test during the load cycle based on the hysteresis loop inflection point in the stress-strain response data, it is specifically used for:

[0038] The stress-strain response data is divided into independent loop blocks according to the load period;

[0039] Locate the curvature abrupt change point in the independent loop block as a candidate inflection point;

[0040] Candidate inflection points that satisfy the condition that the cross product of three consecutive point vectors has the same sign are selected as valid inflection points;

[0041] Based on the effective inflection point, the upper and lower segments of the return line are divided to obtain the peak stress, valley stress, and the area enclosed by the return line of the spring under test during the load cycle.

[0042] In a preferred embodiment, the area enclosed by the loop includes:

[0043] Using the effective inflection point as the boundary, the independent loop block is decomposed into polygonal sub-regions;

[0044] The polygonal sub-region is triangularly divided, and the resulting polygonal sub-regions are summed to obtain the loop-enclosed area of ​​the independent loop block. The formula for calculating the loop-enclosed area is as follows:

[0045]

[0046] In the formula, Let be the area enclosed by the loop. The total number of boundary points of the independent loop block. Let be the boundary point ordinal number of the independent loop block. For the first The coordinates of the boundary points The coordinates of the geometric center point of the loop are... For the first The coordinates of the boundary points This is the vector cross product operator. This is a modulo operation.

[0047] In a preferred embodiment, when the life assessment module generates a life assessment report for the spring under test based on the peak stress, the valley stress, and the area enclosed by the loop, it is specifically used for:

[0048] A stress feature matrix is ​​constructed using the peak stress and the valley stress as row vectors;

[0049] The area enclosed by the loop is converted into an energy dissipation density value and associated with an additional dimension of the stress feature matrix;

[0050] The damage accumulation trajectory curve of the spring under test is obtained by nonlinear fitting of the associated stress feature matrix.

[0051] Based on the number of load cycles corresponding to the inflection point of the damage accumulation trajectory curve, a life assessment report for the spring under test is generated.

[0052] To address the above problems, the present invention also provides a method for dynamic balance evaluation of spring fatigue life, the method comprising:

[0053] S1. Obtain the reference length of the spring to be measured in its free state;

[0054] S2. Apply a stepped increasing preload to the spring under test. When the deformation rate of the spring under test is detected to be less than a set threshold, it is determined that the spring under test has reached a dynamic equilibrium state.

[0055] S3. Based on the vibration spectrum characteristics of the spring under test in the service environment, generate the variable amplitude load timing signal of the spring under test;

[0056] S4. Under the dynamic equilibrium state, load is applied to the spring under test based on the variable amplitude load timing signal, and stress-strain response data of the spring under test are collected simultaneously.

[0057] S5. Identify the peak stress, valley stress and hysteresis loop area of ​​the spring under test during the load cycle based on the inflection point of the hysteresis loop in the stress-strain response data.

[0058] S6. Generate a life assessment report for the spring under test based on the peak stress, the valley stress, and the area enclosed by the loop.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] 1. This invention, by setting up a dynamic equilibrium state determination module and employing a stepped incremental preload method combined with deformation rate judgment, can accurately determine the dynamic equilibrium state of the spring under test. This provides a stable benchmark for subsequent load loading and data acquisition, ensuring that the collected stress-strain response data more closely reflects the actual stress conditions, thereby improving the accuracy of fatigue life assessment. Simultaneously, the working condition spectrum loading module generates a variable amplitude load timing signal based on the vibration spectrum characteristics of the service environment, making the loading process closer to the actual working state of the spring and laying the foundation for accurate assessment.

[0061] 2. This invention identifies key stress parameters and the area enclosed by the hysteresis loop inflection point. Combined with the stress characteristic matrix and damage accumulation trajectory curve constructed by the life assessment module, it can comprehensively capture the damage pattern of the spring. Furthermore, the compensation mechanism during load application effectively reduces the deviation between the actual load and the target load, further improving data reliability. The resulting life assessment report more accurately reflects the fatigue life of the spring, providing strong support for spring performance evaluation. Attached Figure Description

[0062] Figure 1 This is a system architecture diagram of a spring fatigue life testing system provided in an embodiment of the present invention;

[0063] Figure 2 This is a flowchart illustrating a dynamic balance evaluation method for spring fatigue life according to an embodiment of the present invention.

[0064] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0067] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0068] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0069] In practice, the server-side equipment deployed in a spring fatigue life testing system may consist of one or more devices. The aforementioned spring fatigue life testing system can be implemented as: a business instance, a virtual machine, or hardware devices. For example, the spring fatigue life testing system can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, the spring fatigue life testing system can be understood as software deployed on a cloud node to provide spring fatigue life testing services to various user terminals. Alternatively, the spring fatigue life testing system can also be implemented as a virtual machine deployed on one or more devices in a cloud node. This virtual machine contains application software for managing various user terminals. Or, the spring fatigue life testing system can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide spring fatigue life testing services to various user terminals.

[0070] In terms of implementation, the spring fatigue life testing system and the user terminal are mutually compatible. That is, if the spring fatigue life testing system is implemented as an application installed on a cloud service platform, then the user terminal is implemented as a client that establishes a communication connection with the application; or if the spring fatigue life testing system is implemented as a website, then the user terminal is implemented as a webpage; or if the spring fatigue life testing system is implemented as a cloud service platform, then the user terminal is implemented as a mini-program in an instant messaging application.

[0071] like Figure 1 The figure shown is a system architecture diagram of a spring fatigue life testing system provided in an embodiment of the present invention.

[0072] The spring fatigue life testing system 100 of this invention can be located in a cloud server. In terms of implementation, it can be used as one or more service devices, or as an application installed in the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed into a website. Depending on the functions implemented, the spring fatigue life testing system 100 may include a reference length acquisition module 101, a dynamic equilibrium state determination module 102, a working condition spectrum loading module 103, a stress-strain response data acquisition module 104, a damage feature extraction module 105, and a life assessment module 106. The module described in this invention can also be called a unit, referring to a series of computer program segments that can be executed by an electronic device's processor and perform a fixed function, stored in the electronic device's memory.

[0073] In this embodiment of the invention, each of the above-mentioned modules in the spring fatigue life testing system can be implemented independently and can call other modules. Here, "calling" can be understood as one module connecting to multiple modules of another type and providing corresponding services to those connected modules. In the spring fatigue life testing system provided by this embodiment of the invention, the applicable scope of the spring fatigue life testing system architecture can be adjusted by adding modules and directly calling them without modifying the program code, achieving cluster-based horizontal expansion to quickly and flexibly expand the spring fatigue life testing system. In practical applications, the above-mentioned modules can be set in the same device or different devices, or they can be set in a virtual device, such as a service instance in a cloud server.

[0074] The following describes the components and workflow of the spring fatigue life testing system using specific embodiments:

[0075] The reference length acquisition module 101 is used to acquire the reference length of the spring to be tested in a free state;

[0076] In this embodiment of the invention, when the reference length acquisition module acquires the reference length of the spring to be measured in a free state, it is specifically used for:

[0077] The unloaded spring under test is mounted on the base of the test platform;

[0078] The reference length data of the spring under test is generated based on the projected distance between the spring under test and the reference surface in the base of the test platform.

[0079] Specifically, place the test platform base stably on the leveled workbench. Use a spirit level against the upper surface of the base to observe whether the bubble in the spirit level is centered. If the bubble is off-center, adjust the support feet of the workbench until the bubble is centered, ensuring the base is level. Check the circular mounting hole on the base for mounting the spring to be tested. Insert a cylindrical cleaning rod with a diameter slightly smaller than the mounting hole into the hole and wipe it back and forth to remove any residual metal shavings, dust, or other debris. Then, blow compressed air into the mounting hole for 10 seconds to thoroughly clean any impurities in the gaps inside the hole. Remove the unloaded spring to be tested. The two ends of the spring are flat, round faces. Gently wipe both ends with a lint-free cloth dampened with alcohol to remove fingerprints and oil stains. Place it on clean filter paper and let it stand for 2 minutes to allow the alcohol to evaporate completely. Pick up the spring and align the round end of one end with the mounting hole on the base. Keep the spring axis aligned with the axis of the mounting hole. Slowly insert the spring end into the mounting hole to a depth of 1 / 3 of the spring end diameter. At this point, the inner wall of the mounting hole will be in close contact with the spring end. Gently shake the spring with your hand. You should not feel any obvious looseness, and the spring will remain vertical under its own weight without any tilting. Installation is now complete.

[0080] Further, a reference surface is defined at the edge of the test platform base. This reference surface is a rectangular metal plate embedded in the base, with a precision-ground surface and a flatness error not exceeding 0.01 mm. A laser rangefinder is fixed to an adjustable-height bracket, ensuring the laser emitter of the rangefinder is at the same height as the reference surface. The bracket is adjusted so that the laser beam is perpendicular to the reference surface. The rangefinder is then turned on, and the laser beam forms a clear spot on the reference surface. The initial distance value displayed by the rangefinder is recorded; this value represents the distance from the rangefinder to the reference surface. Keeping the positions of the rangefinder and the bracket unchanged, the position of the spring under test is adjusted so that the spring's axis is in the same vertical plane as the laser beam. The laser beam originates from the reference surface, passes through space, and projects onto the end face of the spring away from the base, forming a spot at the center of the end face. After the value displayed by the rangefinder stabilizes, this value is recorded. Subtracting this value from the previously recorded initial distance value yields the projected distance between the spring under test and the reference surface in the test platform base.

[0081] Furthermore, the projected distance between the spring under test and the reference surface in the test platform base is used as the raw data. The installation status of the spring under test is checked again to confirm that the spring is in a naturally extended state without load, and that it has not been subjected to tensile or compressive forces during installation. The axis of the spring remains vertical and does not bend. Since this projected distance precisely corresponds to the straight-line distance between the two ends of the spring under no load and correct installation, this projected distance value is directly determined as the reference length data of the spring under test, and this data is recorded in a dedicated test data table. The data table must indicate the measurement time, test platform number, and spring number.

[0082] In summary, obtaining the reference length of the spring under test in its free state can be achieved by installing the unloaded spring on the test platform base and generating reference length data based on its projected distance from the reference surface in the base. This provides an initial reference for subsequent dynamic equilibrium determination, load loading, and stress-strain response analysis.

[0083] In summary, the accurate acquisition of this reference length ensures that subsequent testing procedures are conducted under a unified initial dimensional reference, reducing interference from initial length measurement deviations on dynamic equilibrium state judgment, load application accuracy, and stress-strain data interpretation. This lays the foundation for the accuracy of the entire spring fatigue life testing process and helps improve the reliability of the final life assessment results.

[0084] The dynamic equilibrium state determination module 102 is used to apply a stepped increasing preload to the spring under test. When the deformation rate of the spring under test is detected to be less than a set threshold, it is determined that the spring under test has reached a dynamic equilibrium state.

[0085] In this embodiment of the invention, when the dynamic equilibrium state determination module applies a stepped increasing preload to the spring under test, and determines that the spring under test has reached a dynamic equilibrium state when the deformation rate of the spring under test is less than a set threshold, it is specifically used for:

[0086] The incremental step size of the stepped load is determined based on the rated load parameters of the spring under test;

[0087] The preload is applied to the spring under test step by step according to the incremental step size;

[0088] During the load holding period, the axial deformation data of the spring under test is collected in real time;

[0089] The real-time deformation rate curve of the spring under test is generated based on the axial deformation data;

[0090] When the real-time deformation rate curve remains below the rate threshold, the spring under test is determined to have reached a dynamic equilibrium state.

[0091] Specifically, obtain the rated load parameter of the spring under test, which is the maximum load value that the spring can withstand, and record it on the spring's product manual or label. Divide the rated load parameter into 10 equal parts, and the value of each part is the increment step of the stepped load. For example, if the rated load parameter is 1000 Newtons, then the increment step is 100 Newtons.

[0092] Furthermore, a loading device is installed on the test platform. This device includes a tray for placing weights, which is connected to the upper end of the spring under test via a rope. Standard weights of corresponding weights are prepared according to the determined increment step, with each weight equal to the increment step value. The first weight is placed on the tray, and its weight is transferred to the spring under test via the rope, subjecting the spring to a load equal to the increment step value. This load is maintained for 5 minutes, completing the first preload application. After 5 minutes, the first weight is removed, and a second weight is placed, increasing the load on the spring by one increment step. This is also maintained for 5 minutes. The number of weights is increased incrementally in this manner, achieving step-by-step preload application to the spring under test according to the increment step.

[0093] Furthermore, a displacement sensor is fixed to both the upper and lower ends of the spring under test. The upper sensor is fixed to the bottom of the loading device's tray, and the lower sensor is fixed to the test platform base. The connection between the two sensors is aligned with the spring's axial direction. The sensors are connected to a data acquisition device, which is set to acquire data once per second. When a preload is applied to the spring and maintained, the sensors monitor the change in distance between the upper and lower ends of the spring in real time. The value of this distance change is the axial deformation data of the spring under test, and the data acquisition device automatically records each acquired axial deformation data.

[0094] Furthermore, the data acquisition instrument is connected to a computer to retrieve the acquired axial deformation data. With time as the horizontal axis and the change in axial deformation data per unit time as the vertical axis, the deformation changes at different time points are marked sequentially on the coordinate system. Then, these marked points are connected by a smooth curve, and the resulting curve is the real-time deformation rate curve of the spring under test.

[0095] Furthermore, a rate threshold is set, which is a fixed value representing the maximum allowable value of the spring deformation rate. The generated real-time deformation rate curve is compared with the rate threshold in real time. When the rate values ​​corresponding to all points on the curve are less than the rate threshold, and this state remains unchanged for 10 minutes, an automatic prompt signal is issued, at which point it is determined that the spring under test has reached a dynamic equilibrium state.

[0096] In summary, by determining the stepped load increment step size based on the rated load parameters of the spring under test, applying preload step by step, and collecting axial deformation data in real time, the deformation change law of the spring under load can be accurately tracked. When the deformation rate remains below the set threshold, a dynamic equilibrium state is determined to be reached, which ensures that the spring is in a stable mechanical state during subsequent load loading, avoiding fluctuations in stress-strain response data caused by initial instability, and providing reliable benchmark conditions for subsequent tests.

[0097] In summary, this method of determining dynamic equilibrium can fully release the residual stress and unstable deformation of the spring during the initial loading stage, making the spring's mechanical properties closer to its stable state during long-term service. Subsequent variable-amplitude load loading and data acquisition based on this state can significantly improve the authenticity and consistency of stress-strain response data, thereby providing more accurate raw data for damage feature extraction and life assessment, and helping to enhance the credibility of the final life assessment report.

[0098] The working condition spectrum loading module 103 is used to generate the variable amplitude load time sequence signal of the spring under test based on the vibration spectrum characteristics of the spring under test in the service environment.

[0099] In this embodiment of the invention, when the operating condition spectrum loading module generates the variable amplitude load time sequence signal of the spring under test based on the vibration spectrum characteristics of the spring under test in the service environment, it is specifically used for:

[0100] Collect vibration acceleration data of the spring under test under typical working conditions;

[0101] The vibration acceleration data is converted into the frequency domain to obtain the main energy distribution frequency band of the spring under test;

[0102] The load amplitude variation range of the spring under test is determined based on the main energy distribution frequency band.

[0103] A random amplitude modulation signal is generated based on the range of load amplitude variation, and a variable amplitude load timing signal of the spring under test is constructed based on the random amplitude modulation signal.

[0104] When the load spectrum loading module generates a random amplitude modulation signal based on the load amplitude variation range and constructs the variable amplitude load timing signal of the spring under test based on the random amplitude modulation signal, it is specifically used for:

[0105] Establish a mapping relationship between the range of load amplitude variation and the time series;

[0106] A basic sinusoidal carrier signal is constructed under the constraints of the aforementioned mapping relationship;

[0107] The amplitude of the basic sinusoidal carrier signal is randomized.

[0108] The randomized base sinusoidal carrier signal is subjected to time-domain smoothing filtering to obtain the variable amplitude load timing signal of the spring under test.

[0109] Specifically, the spring under test is mounted on a vibration test bench simulating typical operating conditions. Typical operating conditions refer to the vibration environments commonly encountered by springs during actual operation, such as the vibration state of an automotive suspension system. A triaxial accelerometer is fixed at the middle of the spring. The sensor is connected to a data acquisition device via wires. The data acquisition device is turned on and the sampling frequency is set to 1000 times per second. The vibration test bench is started to operate according to the vibration parameters of typical operating conditions. At the same time, the data acquisition device begins to record the acceleration values ​​of the spring detected by the sensor during the vibration process. These values ​​are the vibration acceleration data of the spring under test under typical operating conditions. The acquisition time is 30 minutes, during which the operating parameters of the test bench are kept stable.

[0110] Furthermore, select the time-domain waveform corresponding to the vibration acceleration data, click the "Frequency Domain Conversion" function button to process the acceleration data in the time domain, convert the acceleration value at each moment into energy values ​​at different frequencies, and after processing, generate a spectrum diagram with frequency as the horizontal axis and energy value as the vertical axis. From the spectrum diagram, several consecutive frequency intervals with high energy values ​​can be clearly observed. These intervals are the main energy distribution frequency bands of the spring under test.

[0111] Furthermore, based on the obtained main energy distribution frequency band, the load amplitude conversion standard for this type of spring in the corresponding frequency range was consulted. The standard clearly specifies the load amplitude range corresponding to different frequency ranges. Substituting the lowest and highest frequencies in the main energy distribution frequency band into the conversion relationship in the standard, two corresponding load amplitudes were obtained. The lower load amplitude is the lower limit of the range, and the higher load amplitude is the upper limit of the range. The interval formed by these two values ​​is the load amplitude variation range of the spring under test.

[0112] Further, input the determined load amplitude variation range, set the signal modulation mode to random amplitude modulation, and ensure the modulation frequency range remains consistent with the main energy distribution frequency band. Click the "Generate Signal" button to randomly select different amplitudes within the load amplitude variation range, and modulate these amplitudes at random time intervals to form a series of time-varying amplitude signals; these signals are the random amplitude modulated signals. Subsequently, arrange the random amplitude modulated signals in chronological order to form a continuous signal sequence whose amplitude changes continuously over time; this sequence is the variable amplitude load timing signal of the spring under test.

[0113] Specifically, a two-dimensional table is created. The first column of the table is set as the time series, starting at time 0. A time point is recorded every 0.1 seconds, for a total of 1000 time points to form a complete time series. The second column of the table is set as the load amplitude. The determined range of load amplitude variation is decomposed into 1000 specific load amplitude values ​​in a uniform distribution, each value falling within this range. These 1000 load amplitude values ​​are then sequentially filled into the row corresponding to each time point in the time series, ensuring that each time point has a unique corresponding load amplitude. This establishes a mapping relationship between the load amplitude variation range and the time series. This mapping relationship is automatically saved and displayed in a chart format, with the horizontal axis representing the time series and the vertical axis representing the corresponding load amplitude.

[0114] Furthermore, the established mapping relationship is invoked, and the period of the basic sinusoidal carrier signal is determined according to the time series intervals in the mapping relationship, ensuring that the period matches the time interval. The frequency of the basic sinusoidal carrier signal is set, with the frequency value referencing the main vibration frequency of the spring under test under typical operating conditions, ensuring that the signal can reflect the actual force characteristics of the spring. Based on the time series and the corresponding load amplitude variation range, a continuous waveform signal is generated according to the variation law of the sine curve. The variation trend of this signal is constrained by the mapping relationship, that is, the signal amplitude at each time point does not exceed the load amplitude variation range. This signal is the constructed basic sinusoidal carrier signal, and its waveform is displayed in real time.

[0115] Further, prepare a list of 1000 random numbers, with values ​​ranging from 0.8 to 1.2, generated using a random number generator to ensure each value is unique. Import this list of random numbers into signal processing software, and divide the basic sinusoidal carrier signal into 1000 segments according to time intervals, each segment corresponding to the signal amplitude at a specific time point. Then, multiply each random number in the list sequentially by the amplitude of its corresponding signal segment to obtain a new signal amplitude. This causes the signal amplitude at each time point to change randomly from its original value, while the changed amplitude remains within the range of load amplitude variation. This completes the amplitude randomization processing of the basic sinusoidal carrier signal, resulting in an irregularly fluctuating signal waveform.

[0116] Furthermore, the time-domain smoothing filter function is selected. This function processes the signal using a moving average method, setting the sliding window size to 5 time intervals, meaning each window contains the signal amplitude at 5 consecutive time points. Starting from the first time point of the randomized base sinusoidal carrier signal, the average of the 5 signal amplitudes within each sliding window is calculated, and this average is used as the new signal amplitude at the middle time point of the window. This process is repeated for the entire signal, smoothing out sharp waveforms with sudden changes. The resulting signal exhibits continuous and smooth amplitude changes, consistent with the dynamic characteristics of the actual force applied to the spring under test. This signal is the variable amplitude load time-series signal of the spring under test. This signal is saved, and the corresponding time-domain waveform is generated.

[0117] In summary, generating variable amplitude load time-series signals based on the vibration spectrum characteristics of the spring under test in its service environment has significant benefits. By collecting vibration acceleration data under typical working conditions and performing frequency domain transformation, the main energy distribution frequency bands can be accurately captured. The load amplitude variation range determined accordingly is closer to the actual force characteristics of the spring, enabling the generated variable amplitude load time-series signals to realistically simulate the dynamic load action in the service environment.

[0118] In summary, by establishing a mapping relationship between load amplitude and time series, and performing amplitude randomization and time-domain smoothing filtering on the basic sinusoidal carrier signal, the accuracy and continuity of the variable amplitude load time series signal are further improved. Loading based on this ensures that the load on the spring closely matches the actual service conditions, providing reliable input for the accurate acquisition of subsequent stress-strain response data. This lays a solid foundation for the scientific basis of damage feature extraction and life assessment, effectively enhancing the authenticity and effectiveness of fatigue life testing.

[0119] The stress-strain response data acquisition module 104 is used to load the spring under test based on the variable amplitude load time sequence signal under the dynamic equilibrium state, and simultaneously acquire the stress-strain response data of the spring under test.

[0120] In this embodiment of the invention, when the stress-strain response data acquisition module performs load loading on the spring under test based on the amplitude load timing signal under the dynamic equilibrium state, and simultaneously acquires the stress-strain response data of the spring under test, it is specifically used for:

[0121] The spring under test is loaded according to the variable amplitude load timing signal;

[0122] Based on the deviation between the target load value corresponding to the variable amplitude load time sequence signal and the actual load value during the load loading, the compensation value of the load is calculated, wherein the calculation formula of the compensation value is as follows:

[0123]

[0124] In the formula, The compensation value is... This is the proportional gain coefficient. The deviation amount, This is the integral gain coefficient. The time factor;

[0125] The compensation value is superimposed on the driving current command of the load, and the stress-strain response data of the spring under test is collected simultaneously.

[0126] Specifically, the spring to be tested is fixed on a loading test bench, ensuring that the connection at both ends of the spring is secure and that the axis is aligned with the loading direction. The loading test bench is equipped with a loading device that can adjust the output load according to an electrical signal, and this device is connected to a signal generator. The generated variable amplitude load timing signal is imported into the signal generator. The signal generator sends an electrical signal to the loading device according to the load value corresponding to each time point in the timing signal. After receiving the electrical signal, the loading device applies a corresponding load to the spring through its internal power mechanism, and the load magnitude changes with time in accordance with the variable amplitude load timing signal, thereby achieving the loading of the spring to be tested according to the variable amplitude load timing signal.

[0127] Furthermore, a load sensor is installed on the loading device. This sensor is connected to a data acquisition unit and can detect the actual load value during loading in real time and transmit it to the data acquisition unit. Simultaneously, the load value corresponding to each time point in the variable amplitude load timing signal, as the target load value, is also transmitted to the data acquisition unit synchronously. At each time point, the data acquisition unit compares the target load value with the actual load value detected by the sensor, subtracts the actual load value from the target load value, and obtains the load deviation at that time point. Based on the magnitude and direction of the deviation, the load compensation value that needs to be adjusted is determined. The magnitude of the compensation value is equal to the deviation value, and the direction is opposite to the deviation value, thus completing the calculation of the load compensation value.

[0128] Furthermore, the calculated compensation value is transmitted to the control module of the loading device. The control module pre-stores the correspondence between the load drive current command and the load output. After receiving the compensation value, the control module converts the compensation value into a corresponding current adjustment amount according to the correspondence, and then adds this current adjustment amount to the original load drive current command to form a new drive current command, which is then sent to the actuator of the loading device, making the actual load value output by the loading device closer to the target load value. Simultaneously with adjusting the drive current command, stress sensors and strain sensors mounted on the spring surface are activated. The stress sensor detects the magnitude of the stress on the spring, and the strain sensor detects the degree of spring deformation. Both sensors are connected to a data recording instrument, which synchronously collects and stores the stress and strain data output by the sensors at the same time intervals as the amplitude load timing signal. These data together constitute the stress-strain response data of the spring under test.

[0129] Specifically, in the formula for calculating the compensation value, the compensation value is the final calculated value used to adjust the load. The deviation comes from the difference between the target load value corresponding to the variable amplitude load timing signal and the actual load value during loading. The proportional gain coefficient and integral gain coefficient are fixed values ​​preset based on the performance of the loading device, the characteristics of the spring under test, and past test experience. The time factor is the time elapsed from the start of loading to the current moment.

[0130] Furthermore, the significance of this formula lies in calculating a suitable compensation value by comprehensively considering the current deviation and the cumulative deviation over a period of time. The proportional gain coefficient multiplied by the current deviation is used to quickly respond to the current deviation, while the integral gain coefficient multiplied by the cumulative deviation over time is used to eliminate deviations that have existed for a long time. The combined effect of the two allows the calculated compensation value to adjust the load more accurately, making the actual load value closer to the target load value.

[0131] Furthermore, from a trend perspective, when the deviation is positive and remains constant, the cumulative effect of the deviation over time will continuously increase, causing the integral part of the compensation value to gradually increase, while the proportional part remains unchanged, thus the compensation value will gradually increase. When the deviation is negative and remains constant, the cumulative effect of the deviation over time will continuously decrease, causing the integral part of the compensation value to gradually decrease, while the proportional part remains unchanged, thus the compensation value will gradually decrease. When the deviation is zero, the proportional part is zero. If there has been a previous accumulation of deviation, the integral part will maintain its corresponding value based on the accumulation. If there has been no previous accumulation of deviation, the integral part is also zero, and the compensation value is zero.

[0132] In summary, the dynamic equilibrium state provides stable initial mechanical conditions for load loading, avoiding additional deformation interference caused by the spring not reaching a stable state, and ensuring consistency between the load loading process and the actual stress state of the spring. Furthermore, based on the variable amplitude load timing signal, the dynamic stress process of the spring in the service environment can be accurately reproduced, making the stress-strain response data more closely match the mechanical performance under actual working conditions.

[0133] In summary, the synchronous acquisition mechanism ensures the temporal correlation between stress-strain response data and the load application process, guaranteeing the timeliness and relevance of the data. Furthermore, the compensation formula used during load application corrects for deviations between the actual and target loads, further improving the accuracy of load application. This results in more accurate and reliable acquired stress-strain response data, providing high-quality raw data support for subsequent damage feature extraction and life assessment, and ultimately enhancing the accuracy of fatigue life assessment.

[0134] The damage feature extraction module 105 is used to identify the peak stress, valley stress and hysteresis loop area of ​​the spring under test in the load cycle based on the hysteresis loop inflection point in the stress-strain response data.

[0135] In this embodiment of the invention, when the damage feature extraction module identifies the peak stress, valley stress, and hysteresis loop enclosed area of ​​the spring under test during the load cycle based on the hysteresis loop inflection point in the stress-strain response data, it is specifically used for:

[0136] The stress-strain response data is divided into independent loop blocks according to the load period;

[0137] Locate the curvature abrupt change point in the independent loop block as a candidate inflection point;

[0138] Candidate inflection points that satisfy the condition that the cross product of three consecutive point vectors has the same sign are selected as valid inflection points;

[0139] Based on the effective inflection point, the upper and lower segments of the return line are divided to obtain the peak stress, valley stress, and the area enclosed by the return line of the spring under test during the load cycle.

[0140] The area enclosed by the loop includes:

[0141] Using the effective inflection point as the boundary, the independent loop block is decomposed into polygonal sub-regions;

[0142] The polygonal sub-region is triangularly divided, and the resulting polygonal sub-regions are summed to obtain the loop-enclosed area of ​​the independent loop block. The formula for calculating the loop-enclosed area is as follows:

[0143]

[0144] In the formula, Let be the area enclosed by the loop. The total number of boundary points of the independent loop block. Let be the boundary point ordinal number of the independent loop block. For the first The coordinates of the boundary points The coordinates of the geometric center point of the loop are... For the first The coordinates of the boundary points This is the vector cross product operator. This is a modulo operation.

[0145] Specifically, the stress-strain response data of the spring under test is imported. This data is a set of stress values ​​and corresponding strain values ​​recorded in chronological order. Simultaneously, the variable amplitude load time sequence signal is imported, from which the load period is extracted—the time it takes for the load to change from its initial state back to the same initial state. Segmentation parameters are set to divide the stress-strain response data into multiple continuous segments with the load period as the interval. Each segment contains all the stress-strain data within one load period. These segments are called independent loop blocks, and each independent loop block is presented as a separate curve, with strain on the horizontal axis and stress on the vertical axis.

[0146] Furthermore, the curvature analysis tool is invoked to analyze the stress-strain curve of each independent loop block. The tool calculates the degree of curvature of the curve at each point. When the curvature of the curve suddenly changes significantly from a certain point, that is, when the curvature value increases or decreases significantly, this point is the curvature abrupt change point. All such points are marked on the curve and identified as candidate inflection points. Each candidate inflection point corresponds to a specific set of stress and strain values.

[0147] Furthermore, each candidate inflection point is processed by taking each candidate inflection point and its two adjacent points (three consecutive points in total). These three points are labeled as point 1, point 2, and point 3 according to their order on the curve. The vectors formed from point 1 to point 2 and from point 2 to point 3 are calculated, and then the cross product of these two vectors is calculated. The result of the cross product will be either positive or negative. If the cross product of the three consecutive points has the same sign (i.e., all are positive or all are negative), then the candidate inflection point is selected as a valid inflection point. The labels of these valid inflection points are retained, and the labels of candidate inflection points that do not meet the criteria are removed.

[0148] Furthermore, based on the position of the effective inflection point on the stress-strain curve of the independent loop block, the curve is divided into an upward segment and a downward segment. The upward segment refers to the part extending from the starting point of the curve through the effective inflection point in the direction of increasing stress, while the downward segment refers to the part extending from the effective inflection point in the direction of decreasing stress to the end point of the curve. In the divided upward segment, the point with the maximum stress value is found; the stress value corresponding to this point is the peak stress of the spring under test during the load cycle. In the downward segment, the point with the minimum stress value is found; the stress value corresponding to this point is the valley stress. The area of ​​the closed region enclosed by the upward and downward segments of the stress-strain curve is calculated, and the resulting area is the loop enclosed area.

[0149] Specifically, retrieve the stress-strain curves of the independent loop blocks with marked effective inflection points, and arrange them sequentially according to the order in which the effective inflection points appear on the curves. Starting from the first effective inflection point, connect the inflection point with the next adjacent effective inflection point with a straight line, and continue connecting the next adjacent effective inflection point until the last effective inflection point is connected to the first effective inflection point with a straight line, forming a closed dividing line. These dividing lines and the curve segments of the independent loop blocks together enclose multiple continuous and non-overlapping regions. Each region consists of three or more sides, which is a polygonal sub-region. Each polygonal sub-region is marked with a different color for easy distinction.

[0150] Further, in a selected polygonal sub-region, a vertex of that region is designated as a reference point. Starting from the reference point, straight lines are drawn to all non-adjacent vertices of the polygon. These lines divide the polygon into multiple triangles, where each triangle's three vertices are vertices of the original polygon, and there is no overlap between triangles. This process is repeated for all polygonal sub-regions to obtain all triangles. For each triangle, one side is selected as the base, and its length is measured. The vertical distance from the vertex opposite the base to the base is then measured as the height. The base length is multiplied by the height, and the result is divided into two equal parts to obtain the area of ​​the triangle. The areas of all triangles within the same polygonal sub-region are summed to obtain the area of ​​the polygonal sub-region. Finally, the areas of all polygonal sub-regions are summed, and the total sum is the area enclosed by the loops of the independent loop blocks. This area value is recorded and displayed in the results panel.

[0151] Specifically, the area enclosed by the loop is the total area of ​​the region enclosed by the calculated boundaries of the independent loop blocks. The total number of boundary points of the independent loop blocks refers to the number of all points constituting the boundaries of the independent loop blocks. These points come from the stress-strain curves of the independent loop blocks, including effective inflection points and other characteristic points on the curves. The boundary point ordinal number is the sequential numbering of these boundary points, starting from 1 and increasing sequentially. The coordinates of the i-th boundary point and the (i+1)-th boundary point are the specific position values ​​of each boundary point in a coordinate system with strain as the horizontal axis and stress as the vertical axis, obtained by recording stress-strain response data. The coordinates of the geometric center point of the loop are the position values ​​of the point obtained by averaging the horizontal coordinates of all boundary points as the x-coordinate and averaging the vertical coordinates of all boundary points as the y-coordinate.

[0152] Furthermore, the significance of this formula lies in the fact that by calculating the magnitude of the cross product of the vectors formed by the geometric center point of the loop and two adjacent boundary points, and then taking half of the sum of all these magnitudes, the loop-enclosed area of ​​the independent loop block can be obtained. Here, the magnitude of the cross product of the vectors formed by each adjacent boundary point and the geometric center point corresponds to the area of ​​a triangle. By summing these triangle areas and taking half, the area of ​​the closed region enclosed by all boundary points can be accurately calculated, that is, the loop-enclosed area.

[0153] Furthermore, from a trend perspective, when the total number of boundary points of independent loop blocks increases and the distribution range of boundary points expands (i.e., the distance between adjacent boundary points increases), the magnitude of each vector cross product will increase accordingly, and the sum of all magnitudes will also increase, resulting in an increasing loop enclosed area. Conversely, when the total number of boundary points decreases or the distribution range of boundary points shrinks (i.e., the distance between adjacent boundary points decreases), the magnitude of each vector cross product will decrease accordingly, and the sum of all magnitudes will also decrease, resulting in a decreasing loop enclosed area. When the distribution of boundary points remains unchanged, but the total number of boundary points increases proportionally, and the newly added points are evenly distributed among the original adjacent boundary points, the sum of the magnitudes of the vector cross products does not change significantly, and the loop enclosed area remains basically stable.

[0154] In summary, by dividing the stress-strain response data into independent loop blocks according to the load cycle, locating curvature change points as candidate inflection points, and screening for effective inflection points that satisfy the condition that the cross product of three consecutive points has the same sign, it is possible to accurately divide the upward and downward segments of the loop, thereby accurately identifying peak stress and valley stress, and providing a reliable basis for capturing the extreme stress state of the spring in the load cycle.

[0155] In summary, by decomposing independent loop blocks into polygonal sub-regions using effective inflection points as boundaries, and calculating the area enclosed by the loops through triangular subdivision and cumulative calculation, the energy dissipation of the spring during load cycles can be accurately quantified. This process not only ensures the accuracy and completeness of damage feature parameter extraction but also provides high-quality data support for subsequent damage accumulation trajectory fitting and life assessment based on these features, contributing to improving the scientific rigor and accuracy of spring fatigue life assessment.

[0156] The life assessment module 106 is used to generate a life assessment report for the spring under test based on the peak stress, the valley stress, and the area enclosed by the loop.

[0157] In this embodiment of the invention, when the life assessment module generates a life assessment report for the spring under test based on the peak stress, the valley stress, and the area enclosed by the loop, it is specifically used for:

[0158] A stress feature matrix is ​​constructed using the peak stress and the valley stress as row vectors;

[0159] The area enclosed by the loop is converted into an energy dissipation density value and associated with an additional dimension of the stress feature matrix;

[0160] The damage accumulation trajectory curve of the spring under test is obtained by nonlinear fitting of the associated stress feature matrix.

[0161] Based on the number of load cycles corresponding to the inflection point of the damage accumulation trajectory curve, a life assessment report for the spring under test is generated.

[0162] Specifically, import the peak and valley stresses obtained for the spring under test in each load cycle, arranged sequentially according to the order of the load cycles. Create a matrix table where each row represents a load cycle. Use the peak stress corresponding to each cycle as the first element of the row and the valley stress as the second element, filling the table sequentially according to the load cycles. This ensures that each row contains both the peak and valley stresses for its corresponding cycle. The resulting table is the stress characteristic matrix, which is automatically saved and displayed as a two-dimensional array.

[0163] Furthermore, the area enclosed by the loop of each independent loop block is extracted, and the cross-sectional area and effective length in the loading direction of the spring under test are measured. The area enclosed by the loop is divided by the product of the cross-sectional area and the effective length. The result is the energy dissipation density value, and each loop enclosed area corresponds to one energy dissipation density value. The constructed stress feature matrix is ​​retrieved, and a new column is added as an additional dimension. The calculated energy dissipation density values ​​are sequentially filled into this column according to the load cycle, so that the stress feature row vector corresponding to each load cycle is associated with the energy dissipation density value, forming a stress feature matrix containing the additional dimension.

[0164] Furthermore, the data in the associated stress characteristic matrix are correlated with the number of load cycles, with the number of load cycles as the horizontal axis and the cumulative damage calculated from the peak stress, valley stress, and energy dissipation density values ​​of each cycle in the matrix as the vertical axis. Based on the data distribution trend, a smooth curve is automatically generated. By adjusting the curvature of the curve, it is made to closely approximate all data points, ensuring that the curve reflects the overall trend of the cumulative damage as a function of the number of load cycles. This curve is the damage accumulation trajectory curve of the spring under test, which is displayed in a graph with key data points marked.

[0165] Furthermore, zoom in on the damage accumulation trajectory curve and carefully observe its changing trend. When the curve suddenly becomes steep after a gentle rise, or suddenly becomes flat after a steep rise, this point of change is the inflection point. Record the number of load cycles corresponding to the inflection point on the horizontal axis; this number is the life value of the spring under test. Open the report generation tool, input the spring number, test environment parameters, and load cycle characteristics, describe in detail the change process and inflection point characteristics of the damage accumulation trajectory curve, clearly mark the number of load cycles corresponding to the inflection point as the life assessment result, and attach the damage accumulation trajectory curve graph to form a complete life assessment report for the spring under test. After format verification, save the report as an electronic document.

[0166] In summary, by constructing a stress characteristic matrix using peak and valley stresses and converting the area enclosed by the loop into an energy dissipation density value associated with an additional dimension of the matrix, the key mechanical characteristics and energy dissipation information of the spring during load cycles can be comprehensively integrated, forming a multi-dimensional basis for damage assessment. Based on this, nonlinear fitting is performed on the associated stress characteristic matrix, and the resulting damage accumulation trajectory curve accurately reflects the evolution of spring damage with load cycles.

[0167] In summary, generating a life assessment report based on the load cycle count corresponding to the inflection point in the trajectory curve directly correlates the critical damage state of the spring with the actual number of service cycles, making the assessment results more closely reflect the actual fatigue failure process of the spring. This assessment method based on the fusion of multiple characteristic parameters significantly improves the comprehensiveness and accuracy of life assessment, providing a reliable basis for the scientific determination of spring fatigue life.

[0168] Reference Figure 2 The diagram shown is a flowchart illustrating a dynamic balance assessment method for spring fatigue life according to an embodiment of the present invention. In this embodiment, the dynamic balance assessment method for spring fatigue life includes:

[0169] S1. Obtain the reference length of the spring to be measured in its free state;

[0170] S2. Apply a stepped increasing preload to the spring under test. When the deformation rate of the spring under test is detected to be less than a set threshold, it is determined that the spring under test has reached a dynamic equilibrium state.

[0171] S3. Based on the vibration spectrum characteristics of the spring under test in the service environment, generate the variable amplitude load timing signal of the spring under test;

[0172] S4. Under the dynamic equilibrium state, load is applied to the spring under test based on the variable amplitude load timing signal, and stress-strain response data of the spring under test are collected simultaneously.

[0173] S5. Identify the peak stress, valley stress and hysteresis loop area of ​​the spring under test during the load cycle based on the inflection point of the hysteresis loop in the stress-strain response data.

[0174] S6. Generate a life assessment report for the spring under test based on the peak stress, the valley stress, and the area enclosed by the loop.

[0175] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0176] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A spring fatigue life testing system, characterized in that, The system includes a reference length acquisition module, a dynamic equilibrium state determination module, a load condition spectrum loading module, a stress-strain response data acquisition module, a damage feature extraction module, and a life assessment module, wherein: The reference length acquisition module is used to acquire the reference length of the spring to be tested in a free state; The dynamic equilibrium state determination module is used to apply a stepped increasing preload to the spring under test. When the deformation rate of the spring under test is detected to be less than a set threshold, it is determined that the spring under test has reached a dynamic equilibrium state. The working condition spectrum loading module is used to generate the variable amplitude load time sequence signal of the spring under test based on the vibration spectrum characteristics of the spring under test in the service environment. The stress-strain response data acquisition module is used to load the spring under test based on the variable amplitude load time sequence signal under the dynamic equilibrium state, and simultaneously acquire the stress-strain response data of the spring under test. The damage feature extraction module is used to identify the peak stress, valley stress, and hysteresis loop enclosed area of ​​the spring under test during the load cycle based on the hysteresis loop inflection point in the stress-strain response data, including: The stress-strain response data is divided into independent loop blocks according to the load period; Locate the curvature abrupt change point in the independent loop block as a candidate inflection point; Candidate inflection points that satisfy the condition that the cross product of three consecutive point vectors has the same sign are selected as valid inflection points; Based on the effective inflection point, the upper and lower segments of the loop are divided, and the peak stress, valley stress and loop enclosed area of ​​the spring under test in the load cycle are obtained. The life assessment module is used to generate a life assessment report for the spring under test based on the peak stress, the valley stress, and the area enclosed by the loop, including: A stress feature matrix is ​​constructed using the peak stress and the valley stress as row vectors; The area enclosed by the loop is converted into an energy dissipation density value and associated with an additional dimension of the stress feature matrix, wherein the area enclosed by the loop includes: Using the effective inflection point as the boundary, the independent loop block is decomposed into polygonal sub-regions; The polygonal sub-region is triangularly divided, and the resulting polygonal sub-regions are summed to obtain the loop-enclosed area of ​​the independent loop block. The formula for calculating the loop-enclosed area is as follows: ； In the formula, Let be the area enclosed by the loop. The total number of boundary points of the independent loop block. Let be the boundary point ordinal number of the independent loop block. For the first The coordinates of the boundary points The coordinates of the geometric center point of the loop are... For the first The coordinates of the boundary points This is the vector cross product operator. For modulo operation; The damage accumulation trajectory curve of the spring under test is obtained by nonlinear fitting of the associated stress feature matrix. Based on the number of load cycles corresponding to the inflection point of the damage accumulation trajectory curve, a life assessment report for the spring under test is generated.

2. The spring fatigue life testing system as described in claim 1, characterized in that, When the reference length acquisition module acquires the reference length of the spring to be measured in its free state, it is specifically used for: The unloaded spring under test is mounted on the base of the test platform; The reference length data of the spring under test is generated based on the projected distance between the spring under test and the reference surface in the base of the test platform.

3. The spring fatigue life testing system as described in claim 1, characterized in that, When the dynamic equilibrium state determination module applies a stepped increasing preload to the spring under test, and determines that the spring under test has reached a dynamic equilibrium state when the deformation rate of the spring under test is less than a set threshold, it is specifically used for: The incremental step size of the stepped load is determined based on the rated load parameters of the spring under test; The preload is applied to the spring under test step by step according to the incremental step size; During the load holding period, the axial deformation data of the spring under test is collected in real time; The real-time deformation rate curve of the spring under test is generated based on the axial deformation data; When the real-time deformation rate curve remains below the rate threshold, the spring under test is determined to have reached a dynamic equilibrium state.

4. The spring fatigue life testing system as described in claim 1, characterized in that, When the operating condition spectrum loading module generates the variable amplitude load time sequence signal of the spring under test based on the vibration spectrum characteristics of the spring under test in the service environment, it is specifically used for: Collect vibration acceleration data of the spring under test under typical working conditions; The vibration acceleration data is converted into the frequency domain to obtain the main energy distribution frequency band of the spring under test; The load amplitude variation range of the spring under test is determined based on the main energy distribution frequency band. A random amplitude modulation signal is generated based on the range of load amplitude variation, and a variable amplitude load timing signal of the spring under test is constructed based on the random amplitude modulation signal.

5. The spring fatigue life testing system as described in claim 4, characterized in that, When the load spectrum loading module generates a random amplitude modulation signal based on the load amplitude variation range and constructs the variable amplitude load timing signal of the spring under test based on the random amplitude modulation signal, it is specifically used for: Establish a mapping relationship between the range of load amplitude variation and the time series; A basic sinusoidal carrier signal is constructed under the constraints of the aforementioned mapping relationship; The amplitude of the basic sinusoidal carrier signal is randomized. The randomized base sinusoidal carrier signal is subjected to time-domain smoothing filtering to obtain the variable amplitude load timing signal of the spring under test.

6. The spring fatigue life testing system as described in claim 1, characterized in that, When the stress-strain response data acquisition module performs load loading on the spring under test based on the amplitude load time sequence signal under the dynamic equilibrium state, and simultaneously acquires the stress-strain response data of the spring under test, it is specifically used for: The spring under test is loaded according to the variable amplitude load timing signal; Based on the deviation between the target load value corresponding to the variable amplitude load time sequence signal and the actual load value during the load loading, the compensation value of the load is calculated, wherein the calculation formula of the compensation value is as follows: ； In the formula, The compensation value is... This is the proportional gain coefficient. The deviation amount, This is the integral gain coefficient. The time factor; The compensation value is superimposed on the driving current command of the load, and the stress-strain response data of the spring under test is collected simultaneously.

7. A method for dynamic balance evaluation of spring fatigue life, characterized in that, The method for implementing the spring fatigue life testing system according to claim 1 includes: S1. Obtain the reference length of the spring to be measured in its free state; S2. Apply a stepped increasing preload to the spring under test. When the deformation rate of the spring under test is detected to be less than a set threshold, it is determined that the spring under test has reached a dynamic equilibrium state. S3. Based on the vibration spectrum characteristics of the spring under test in the service environment, generate the variable amplitude load timing signal of the spring under test; S4. Under the dynamic equilibrium state, load is applied to the spring under test based on the variable amplitude load timing signal, and stress-strain response data of the spring under test are collected simultaneously. S5. Identify the peak stress, valley stress and hysteresis loop area of ​​the spring under test during the load cycle based on the inflection point of the hysteresis loop in the stress-strain response data. S6. Generate a life assessment report for the spring under test based on the peak stress, the valley stress, and the area enclosed by the loop.

Images