Method and device for evaluating stress of rotating part

Abstract

The invention discloses a rotating part stress evaluation method and device. The method comprises the following steps: acquiring a stress spectrum time domain signal and an acceleration signal of the rotating part in the same bearing environment and in the same sampling time period; according to a corresponding relation between a preset acceleration root-mean-square value and the equivalent stress, the stress spectrum time domain signal of the rotating part and the acceleration signal of the rotating part, constructing a rotating part stress evaluation model in the current bearing environment; and evaluating the stress state of the rotating part in the non-stress sampling time according to the rotating part stress evaluation model in the current bearing environment. According to the technical scheme provided by the invention, the correlation between the acceleration and the stress of the rotating part is constructed, the stress state of the rotating part when the rotating part is difficult to monitor is effectively evaluated through convenient acceleration signal monitoring, and the technical scheme provided by the invention has universality.

Description

Technical Field

[0001] This application relates to the field of stress measurement, and more specifically, to a method and apparatus for stress assessment of rotating components. Background Technology

[0002] Rotating components (such as axles) play a vital role in various industries (such as transportation), and their operating efficiency and structural integrity directly affect the overall performance and safety level of the system in which they are located.

[0003] Conventional stress monitoring methods often rely on contact sensors or destructive testing. However, these methods have significant technical limitations in practical engineering applications, such as cumbersome installation processes, high costs, and the inability to achieve real-time monitoring. Although existing non-destructive evaluation (NDE) technology has formed a relatively mature research system in the field of translational structural components, for rotating components, commonly used destructive testing methods, due to their complex operating conditions and testing difficulties, greatly limit the implementation of real-time operation and maintenance monitoring.

[0004] In this technical field, acceleration monitoring of rotating components is far superior to direct stress measurement in terms of ease of operation. Therefore, designing an efficient, convenient, and accurate stress assessment method for rotating components has become an urgent problem to be solved in this field. Summary of the Invention

[0005] In view of this, in a first aspect, this application proposes a method for stress assessment of rotating components, comprising:

[0006] A method for stress assessment of a rotating component, characterized in that the method includes:

[0007] Acquire the time-domain stress spectrum signal and acceleration signal of the rotating component under the same load environment and within the same sampling time period;

[0008] Based on the preset correspondence between the root mean square value of acceleration and the equivalent stress, the time domain signal of the stress spectrum of the rotating component and the acceleration signal of the rotating component, a stress evaluation model for the rotating component under the current load-bearing environment is constructed.

[0009] Based on the stress assessment model of the rotating component under the current load-bearing environment, the stress state of the rotating component under non-stress sampling time is evaluated.

[0010] Preferably, the expression for the relationship between the preset root mean square value of acceleration and the equivalent stress is:

[0011] σ eq =a RMSk+1 ·10 b ;

[0012] Where, σ eq a represents the equivalent stress across the cross section of the rotating component. RMS denoted by , where is the root mean square value of the acceleration of the rotating component, and k and b are both fitting parameters of the stress evaluation model for the rotating component.

[0013] More preferably, based on the preset correspondence between the root mean square value of acceleration and the equivalent stress, the time-domain signal of the stress spectrum of the rotating component, and the acceleration signal of the rotating component, a stress assessment model for the rotating component under the current load-bearing environment is constructed, including:

[0014] Based on the time-domain signal of the stress spectrum of the rotating component, the equivalent stress of the cross-section of the rotating component is calculated, and

[0015] The root mean square value of the acceleration of the rotating component is calculated based on the acceleration signal of the rotating component.

[0016] The equivalent stress of multiple sets of rotating component cross sections and the root mean square value of the acceleration of the rotating component are substituted into the preset correspondence between the root mean square value of acceleration and the equivalent stress to obtain the fitting curve;

[0017] Based on the fitted curve, a stress assessment model for rotating components under the current load-bearing environment is constructed.

[0018] More preferably, the equivalent stress of the cross-section of the rotating component is calculated based on the time-domain signal of the stress spectrum of the rotating component, including:

[0019] The stress spectrum time-domain signal of the rotating component is analyzed to obtain the stress amplitude of each stress level in the stress spectrum of the rotating component, the number of cycles at each stress level in the stress spectrum of the rotating component, the total number of cycles in the stress spectrum of the rotating component, and the parameters of the SN curve of the rotating component.

[0020] The equivalent stress of the rotating component cross section is calculated based on the stress amplitude of each stress level in the stress spectrum of the rotating component, the number of cycles at each stress level in the stress spectrum of the rotating component, the total number of cycles in the stress spectrum of the rotating component, and the parameters of the SN curve of the rotating component.

[0021] More preferably, the root mean square value of the acceleration of the rotating component is calculated based on the acceleration signal of the rotating component, including:

[0022] The acceleration signal of the rotating component is analyzed to obtain the acceleration value of each sampling point within the sampling time period and the number of acceleration signals collected within the sampling time period;

[0023] The root mean square value of the acceleration of the rotating component is calculated based on the acceleration value of each sampling point within the sampling time period and the number of acceleration signals collected within the sampling time period.

[0024] Preferably, after constructing a stress assessment model for the rotating component in the current load-bearing environment, the method further includes:

[0025] Verify whether the goodness of fit of the stress assessment model for rotating components under the current load-bearing environment meets the preset requirements;

[0026] If satisfied, the stress state of the rotating component under non-stress sampling time is evaluated according to the stress evaluation model of the rotating component under the current load environment.

[0027] If the requirements are not met, the fitting parameters are reselected until the goodness of fit of the stress evaluation model of the rotating component meets the preset requirements.

[0028] Preferably, the stress state of the rotating component under non-stress sampling time is evaluated according to the stress evaluation model of the rotating component under the current load environment, including:

[0029] Acquire the acceleration signal of the rotating component during the non-stress sampling time;

[0030] Based on the acceleration signal of the rotating component during the non-stress sampling time and the stress evaluation model of the rotating component, the time-domain information of the corresponding rotating component stress is obtained;

[0031] Based on the time-domain information of the stress of the rotating component, the stress state of the rotating component under non-stress sampling time is determined.

[0032] More preferably, based on the acceleration signal of the rotating component during the non-stress sampling time and the stress evaluation model of the rotating component, the time-domain information of the corresponding rotating component stress is obtained, including:

[0033] The acceleration signal of the rotating component under non-stress sampling time is divided according to the rotation period of the wheelset system under the current load environment to obtain multiple segments of acceleration signal under non-stress sampling time.

[0034] The acceleration signals under the multiple non-stress sampling times are fed into the stress evaluation model of the rotating component to obtain a discrete equivalent stress sequence of the rotating component;

[0035] The time-domain information of the stress of the discrete rotating component is obtained by reconstructing the equivalent stress sequence in the time domain.

[0036] In a second aspect, embodiments of the present invention also provide a stress evaluation device for rotating components, comprising:

[0037] The signal acquisition module is configured to acquire the stress spectrum time-domain signal and acceleration signal of the rotating component under the same load environment and within the same sampling time period;

[0038] The fitting module is configured to construct a stress evaluation model of the rotating component under the current load environment based on the preset correspondence between the root mean square value of acceleration and the equivalent stress, the time domain signal of the stress spectrum of the rotating component, and the acceleration signal of the rotating component.

[0039] The evaluation module is configured to evaluate the stress state of the rotating component under non-stress sampling time based on the rotating component stress evaluation model under the current load environment.

[0040] Preferably, the expression for the relationship between the preset root mean square value of acceleration and the equivalent stress is:

[0041] σ eq =a RMS k+1 ·10 b ;

[0042] Where, σ eq a represents the equivalent stress across the cross section of the rotating component. RMS denoted by , where is the root mean square value of the acceleration of the rotating component, and k and b are both fitting parameters of the stress evaluation model for the rotating component.

[0043] The stress assessment method for rotating components provided in this application constructs a stress assessment model for rotating components under the current load environment based on the preset correspondence between the root mean square value of acceleration and equivalent stress, the time-domain signal of the stress spectrum of the rotating component, and the acceleration signal of the rotating component. Then, the model is used to assess the stress state of the rotating component under non-stress sampling time. This application's technical solution successfully establishes the correlation between the acceleration and stress of the rotating component. Through relatively convenient acceleration signal monitoring, it effectively assesses the stress state of the rotating component when it is difficult to monitor. Moreover, the technical solution of this application has universality, providing a convenient, real-time, and non-destructive technical solution for the health monitoring and fault diagnosis of rotating components.

[0044] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0045] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application, and the illustrative embodiments and descriptions thereof are used to explain this application. In the drawings:

[0046] Figure 1 A flowchart of the stress assessment method for rotating components according to the preferred embodiment of the application;

[0047] Figure 2 This is a schematic diagram showing the location of the measuring points on the axle section of the preferred embodiment.

[0048] Figure 3 A schematic diagram of the axle stress spectrum for the preferred embodiment of the application;

[0049] Figure 4 A schematic diagram of the acceleration signal of the rotating component according to the preferred embodiment of the application;

[0050] Figure 5 A schematic diagram of the equivalent stress prediction results for a certain route in the preferred embodiment of the application;

[0051] Figure 6 A schematic diagram illustrating the process of reducing the unit segmentation of the root mean square acceleration and axle equivalent stress according to the preferred embodiment of the application;

[0052] Figure 7 A schematic diagram of fitting parameters for the preferred embodiment of the application;

[0053] Figure 8 A schematic diagram showing the predicted stress spectrum of axles on a certain railway line according to the preferred embodiment of the application.

[0054] Figure 9 This is a schematic diagram of a rotating component stress evaluation device according to a preferred embodiment of the application. Detailed Implementation

[0055] The technical solution of this application will now be described in detail with reference to the accompanying drawings and embodiments. To better understand this application, the terms used in this application will first be explained:

[0056] Stress: When an object deforms due to external factors (force, humidity, temperature field changes, etc.), internal forces are generated between the different parts of the object. The internal force per unit area is called stress.

[0057] Stress in rotating components: refers to the internal stress caused by the loads borne by rotating components under various working conditions.

[0058] Equivalent stress of rotating components: This describes the overall stress level at a point or within a region. When a rotating component is subjected to various variable amplitude loads, the actual complex stress state is converted into an equivalent stress amplitude under uniaxial cyclic loading to facilitate fatigue life analysis and load spectrum simplification. This equivalent stress characterizes a stress scalar that has an equivalent fatigue damage effect to the original stress state under the Miner linear cumulative damage criterion, facilitating the standardization of fatigue life calculations under different loading conditions.

[0059] Wheelset: consists of a rotating wheel and a rotating shaft. The rotating wheel is assembled with the rotating shaft by hot pressing or cold pressing to form a wheelset.

[0060] Wheelset rotation cycle: refers to the time required for a wheelset to complete one full rotation.

[0061] SN curve: The SN curve is a curve that uses the fatigue strength of a standard specimen as the vertical axis and the logarithm of the fatigue life as the horizontal axis. It represents the relationship between the fatigue strength and fatigue life of a standard specimen under certain cyclic characteristics. It is also called the stress-life curve.

[0062] Stress spectrum of rotating shaft: refers to the variation of stress on the shaft of a rotating component during operation with time or number of cycles. It is a spectrum of stress amplitudes distributed according to a certain pattern obtained by actual measurement or simulation, reflecting the complex load conditions that the shaft bears under actual working conditions.

[0063] The time-domain signal of the stress spectrum of the rotating shaft is a data sequence obtained directly from sensors, showing the change of stress at different locations of the rotating shaft component over time during operation. Plotted on the horizontal axis (time) and the vertical axis (stress value), it intuitively reflects the instantaneous stress state of the rotating shaft component under dynamic loads.

[0064] Fitting, also known as data fitting or curve fitting, is a method of using mathematical methods to substitute existing data into a numerical expression. When given discrete data, the process of finding a continuous function (i.e., a curve) or a more dense discrete equation that matches the known data is called fitting.

[0065] Fitting parameters: These are unknown constants that need to be determined during function fitting to describe the quantitative relationship between variables.

[0066] Goodness of fit: This measures how well a model fits the data. It represents the model's ability to interpret the data, i.e., how well the model fits the data.

[0067] Confidence level refers to the probability that a population parameter value falls within a certain interval of the sample statistics; that is, the degree of confidence that the population parameter value falls within a certain interval of the sample statistics. It is an indicator used to measure the reliability of statistical inference and is expressed as a percentage.

[0068] The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0069] like Figure 1 As shown, the stress assessment method for rotating components provided in this embodiment includes steps 110-130:

[0070] Step 110: Obtain the time-domain stress spectrum signal and acceleration signal of the rotating component under the same load environment and within the same sampling time period;

[0071] Specifically, the stress spectrum time-domain signal and acceleration signal of the rotating component obtained in this application need to correspond to the same load-bearing environment and the same sampling time period to ensure a relationship between them. The same load-bearing environment can be understood as the rotating component being under the same operating and stress state, and the same sampling time period can be understood as the time period during which the stress spectrum time-domain signal and acceleration signal of the rotating component are sampled. It should be noted that sampling time lengths in minutes or hours are acceptable for subsequent steps, and this application does not limit this.

[0072] In one specific embodiment, the rotating component is an axle, and the bearing environment corresponding to the axle is the vehicle's driving path environment. The sampling time intervals for the axle stress spectrum time-domain signal and the axle acceleration signal are set according to the test conditions. Furthermore, the time-domain signal of the axle stress spectrum, i.e., the time-domain signal of the rotating component's stress spectrum, can be acquired in real time using a stress testing system installed on the axle. Simultaneously, an acceleration sensor is installed on the axle box to acquire the axle box's acceleration signal in real time, i.e., the acceleration signal of the rotating component, and this acceleration signal is preferably a triaxial vibration acceleration signal.

[0073] Step 120: Based on the preset correspondence between the root mean square value of acceleration and the equivalent stress, the time domain signal of the stress spectrum of the rotating component and the acceleration signal of the rotating component, construct a stress assessment model of the rotating component in the current bearing environment.

[0074] Specifically, in the field of mechanics, the greater the change in the rotational speed of a rotating component, the greater the stress the component experiences. Based on this, this application proposes a scheme that uses two concrete indicators—root mean square acceleration and equivalent stress—to intuitively express and quantify the interaction between the acceleration and stress of a rotating component, and uses this as the basis for constructing a rotational stress assessment model.

[0075] In a specific embodiment, the expression for the correspondence between the root mean square value of acceleration and the equivalent stress preset in this application is as shown in equation (1):

[0076] σ eq =a RMS k+1 ·10 b (1)

[0077] Where, σ eq a represents the equivalent stress across the cross section of the rotating component. RMS denoted as the root mean square value of the acceleration of the rotating component, and k and b are both fitting parameters in the stress evaluation model of the rotating component.

[0078] Understandably, equation (1) expresses the fitting function of stress and acceleration proposed in this application, which is also a fitting curve. A digital model can be constructed based on this fitting curve to represent it. In this fitting curve, when using double logarithmic coordinates, k represents the slope and b represents the intercept. There must be some relationship between the acceleration and stress of a rotating component. Equation (1) above is the preferred expression for the correspondence between the root mean square value of acceleration and stress, which was independently constructed by the inventors of this application based on long-term experiments. It can be used for various rotating components.

[0079] Based on Equation (1) above, this application also needs to calculate the equivalent stress of multiple rotating component sections based on the stress spectrum time-domain signal of the rotating component over a period of time, and calculate the root mean square value of multiple rotating component accelerations based on the acceleration signals of multiple rotating components in the same period of time. Then, the calculated equivalent stress values ​​of multiple rotating component sections and the root mean square value of multiple rotating component accelerations are substituted into the preset correspondence between the root mean square value of acceleration and equivalent stress (i.e., Equation 1 above) for fitting, and a fitting curve can be obtained. Based on the fitting curve, a stress evaluation model of the rotating component in the current bearing environment can be constructed.

[0080] Understandably, different load-bearing environments may correspond to different stress assessment models for rotating components. This is because the time-domain stress spectrum and acceleration signals of the rotating component obtained under different load-bearing environments may differ, leading to variations in the equivalent stress and root mean square (RMS) acceleration values ​​of the rotating component's cross-section. Although the same expression relating RMS acceleration to equivalent stress is used, the different parameters result in different fitting curves and thus different stress assessment models. For example, when the rotating component is specifically an axle, its operation and stress state will vary depending on driving requirements and road conditions under different driving environments. Therefore, even for the same axle, the corresponding stress assessment model will differ under different driving environments.

[0081] In a specific embodiment, the calculation of the equivalent stress of the rotating component cross section can be as follows: Analyze the time-domain signal of the stress spectrum of the rotating component to obtain the stress amplitude of each stress level, the number of cycles at each stress level, the total number of cycles in the stress spectrum, and the parameters of the SN curve of the rotating component. Then, based on the stress amplitude of each stress level, the number of cycles at each stress level, the total number of cycles in the stress spectrum, and the parameters of the SN curve, the equivalent stress of the rotating component cross section is calculated. The specific calculation expression is shown in equation (2):

[0082]

[0083] Where, σ eq σ represents the equivalent stress across the cross section of the rotating component. i N represents the stress amplitude at each stress level in the stress spectrum of the rotating component. i The number of cycles for each stress level in the stress spectrum of the rotating component is represented by , N represents the total number of cycles in the stress spectrum of the rotating component, and m represents the parameter of the SN curve of the rotating component.

[0084] In a specific embodiment, the calculation of the root mean square (RMS) value of the acceleration of the rotating component can be performed as follows: The acceleration signal of the rotating component is analyzed to obtain the acceleration value of each sampling point within the sampling time period and the number of acceleration signals collected within the sampling time period. Then, based on the acceleration value of each sampling point within the sampling time period and the number of acceleration signals collected within the sampling time period, the RMS value of the acceleration of the rotating component is calculated. The specific calculation expression is shown in equation (3):

[0085]

[0086] Among them, a RMS a represents the root mean square value of the acceleration of the rotating component. i Let n represent the acceleration value at each sampling point within the sampling time period, where n represents the number of acceleration signals acquired within that time period. Here, the sampling point in the acceleration test refers to the discrete data points on the time axis where the sensor measures the acceleration signal at fixed time intervals (sampling period). For example, if the sensor's sampling frequency is 5000Hz, then 5000 data points are acquired per second.

[0087] To better understand the relationship between the root mean square value of acceleration and the equivalent stress expressed in equation (1) of this application, the equivalent stress σ of the rotating component cross section is used as a basis. eq and the root mean square value of the acceleration of the rotating component a RMS The derivation process of equation (1) above is explained as follows:

[0088] In general, this application uses the least squares method to fit the preset root mean square value of acceleration and the equivalent stress to obtain the corresponding relationship.

[0089] First, based on the equivalent stress σ of the rotating component cross section eq and the root mean square value of the acceleration of the rotating component a RMS The ratio between the two can be calculated, which is referred to in this application as the equivalent surface density ρ, and its calculation expression is shown in equation (4):

[0090]

[0091] To clarify the physical meaning of the equivalent surface density ρ, this application uses a simple harmonic system for qualitative analysis: Assuming a simple harmonic system with amplitude A is used for simple harmonic motion, the total energy E of the system can be expressed as equation (5):

[0092]

[0093] Where K represents the stiffness of the simple resonant system.

[0094] The equation for the vibration acceleration 'a' of the rotating component in a simple harmonic system during motion is given by equation (6):

[0095] a=―ω 2 Asin(ωt+φ) (6)

[0096] Where φ represents the phase difference (or phase angle), which depends on the initial phase and the position of the measurement point, t represents time, ω represents the angular frequency, and the formula for calculating ω based on the period or frequency is ω = 2πf, where f represents the vibration frequency (Hz). The formula for calculating ω based on system parameters is... M represents the support mass, which usually refers to the mass of the vibrating mass block in the system that is supported by elastic elements (such as springs) or support structures, that is, the mass of the object participating in the vibration.

[0097] Based on equation (6), a can be derived. RMS The calculation is as shown in equation (7).

[0098]

[0099] Therefore, the amplitude A can be determined as shown in equation (8).

[0100]

[0101] Substituting equation (8) into equation (5), we obtain equation (9).

[0102]

[0103] Furthermore, assuming M = ρS, where S represents the bearing area of ​​the rotating component shaft, then we have equation (10):

[0104]

[0105] As can be seen from equation (10), for a constant input energy E, the larger the supporting mass M, the smaller the corresponding root mean square value of acceleration. Therefore, the expression for the equivalent surface density ρ can be obtained as shown in equation (11):

[0106]

[0107] Equation (12) can be further derived from equation (11):

[0108]

[0109] Based on this, this application defines a root mean square value 'a' for the acceleration of the rotating component. RMS The logarithmic linear relationship between the surface density ρ and the equivalent surface density ρ is given by expression (13):

[0110] lg(ρ)=k·lga RMS +b+γ (13)

[0111] Where γ represents the error term in the fitting process, and this γ follows a normal distribution N(0,σ). γ 2 ), σ γ The standard deviation of the error term γ, that is, the root mean square value of the acceleration of the rotating component in this application, is represented by α. RMS The relationship between the equivalent surface density ρ and the equivalent surface density ρ is a significant logarithmic linear mapping relationship as shown in equation (13). Substituting equation (4) into equation (13) will lead to equation (1), and the final relationship between the root mean square value of acceleration and the equivalent stress is reflected in equation (1).

[0112] Of course, based on the above derivation, equation (1) also has variations (14) and (15), as follows:

[0113] lg(ρ)=lg(a k ·10 b (14)

[0114] ρ=a k ·10 b (15)

[0115] The above describes the logic for setting the correspondence between the root mean square value of acceleration and the equivalent stress in this application. Regarding the error term γ in the fitting process, this error term represents the difference or deviation between the actual observed value and the model's predicted value. During the fitting process, after fitting the data, a predicted value is obtained. Subtracting the actual observed value from the predicted value yields the error term. If each error term can be considered as the sum of multiple independent, identically distributed random perturbation variables with finite variance, then according to the central limit theorem, as the number of perturbation terms increases, the distribution of the error term will gradually approach a normal distribution. Furthermore, to obtain more conservative and accurate prediction results, it can be assumed that the error term follows a normal distribution. Fitting parameters at different confidence levels are calculated based on the mean and standard deviation of the normal distribution. Using these fitting parameters, more conservative and accurate prediction results can be obtained.

[0116] In practical applications, if it is assumed that the model error term follows a normal distribution and passes relevant verification, then the statistical inferences and predictions based on this model will be more reliable and accurate. Therefore, this application also assumes that the error term γ follows a normal distribution N(0,σ). γ 2 Furthermore, since the error term γ in this application conforms to a normal distribution, a confidence level can be calculated based on this normal distribution. That is, assuming the model's error term γ follows a normal distribution, this distribution property can be used to quantify the uncertainty of the prediction results, thereby calculating the confidence level. The confidence level can be understood as: given sample data and specific statistical assumptions, the constructed confidence interval has the probability of covering the true parameters in a large number of repeated samples at a set proportion (e.g., 95%). This proportion is the confidence level, and its value can be flexibly set according to actual needs, such as the commonly used 90% or 95%. In practical applications, if a confidence threshold (e.g., 90%) is set, then when the confidence interval corresponding to the model's prediction results can cover the observed values, or the confidence level corresponding to the predicted output is not lower than this threshold, it indicates that the model has a certain degree of conservatism and stability under the current conditions, and its prediction results can be considered to have acceptable credibility, and can be used for subsequent engineering analysis or parameter screening processes. In other words, data that meets the confidence level conditions can be considered usable data. For example, if the minimum acceptable confidence level is set to 90%, then if all the predicted data fall within the region with a confidence level greater than 90%, the data is considered usable, and the model can be regarded as an effective model in this application scenario, possessing practical engineering usability.

[0117] Simultaneously, after completing the fitting (obtaining the fitting parameters), a goodness-of-fit index value is obtained, such as the R-squared value of the proportion of variability of explanatory variables in the model. This goodness-of-fit can be used to evaluate the performance of the model. That is, after constructing the stress assessment model of the rotating component under the current load-bearing environment, it is also necessary to verify whether the goodness-of-fit of the stress assessment model of the rotating component under the current load-bearing environment meets the preset requirements. If it does, it means that the current fitting effect is good, and there is no need to adjust the fitting parameters. Then, based on the stress assessment model of the rotating component under the current load-bearing environment, the stress state of the rotating component under non-stress sampling time is evaluated, that is, step 130 is executed; if it does not meet the requirements, it means that the current fitting effect is poor. Then, the fitting parameters k and b are reselected until the goodness-of-fit of the stress assessment model of the rotating component meets the preset requirements.

[0118] Step 130: Based on the stress assessment model of the rotating component under the current load environment, assess the stress state of the rotating component under non-stress sampling time.

[0119] Specifically, stress testing of rotating component shafts is quite complex, requiring extensive preliminary design and preparation. This presents challenges due to the high technical difficulty and cost of monitoring. Furthermore, most current stress tests for rotating component shafts are destructive, making it impossible to directly monitor the stress state of rotating components non-destructively using existing technologies in situations where monitoring is difficult. For example, during high-speed rotation, the rotation period of a rotating shaft is too short, making it impossible to achieve effective non-destructive stress data monitoring within a very short time using current technologies. However, acceleration signals from rotating components are relatively easy to acquire, and the required testing tools and procedures are relatively simple. Therefore, this application adopts a method where, after obtaining a usable rotating component stress assessment model under the current load environment, the model is used to determine the corresponding stress data based on the acceleration data of the rotating component during the non-stress sampling time. This approach assesses the stress state of the rotating component during the non-stress sampling time. By using a simple testing method, this achieves a goal that typically requires complex testing, while simultaneously enabling non-destructive stress detection, making the comprehensive application of rotating component stress monitoring possible. Here, the non-stress sampling time can be understood as the time during which stress data is not (cannot, is not easily) collected from the rotating component.

[0120] In a specific embodiment, the acceleration signal of the rotating component during the non-stress sampling time can be acquired first. Then, based on the acceleration signal of the rotating component during the non-stress sampling time and the stress assessment model of the rotating component, the time-domain information of the corresponding rotating component stress can be obtained. Finally, based on the time-domain information of the rotating component stress, the stress state of the rotating component during the non-stress sampling time can be determined. The specific method for obtaining the time-domain information of the corresponding rotating component stress based on the acceleration signal of the rotating component during the non-stress sampling time and the stress assessment model of the rotating component includes steps 210-230:

[0121] Step 210: Divide the acceleration signal of the rotating component under the non-stress sampling time according to the rotation period of the wheelset system under the current load environment to obtain multiple acceleration signals under the non-stress sampling time.

[0122] Specifically, the data duration of the acceleration signal of the rotating component under non-stress sampling time is the initial duration. The data of this initial duration is divided according to the rotation cycle of the wheelset system under the current load environment, and multiple acceleration signals under non-stress sampling time can be obtained.

[0123] When dividing the data, if the initial duration is greater than the rotation period T of the wheelset system, it is necessary to iteratively compress the data length of the equivalent stress of the initial rotating components until the data segment duration converges to the rotation period T of the wheelset system under the current load conditions. The calculation formula for the rotation period T is as shown in equation (16):

[0124] T=πD / v (16)

[0125] Where D is the diameter of the rolling circle of the rotating component, and v represents the rotational speed of the rotating component.

[0126] Step 220: Input the acceleration signals under the multiple non-stress sampling times into the stress evaluation model of the rotating component to obtain a discrete equivalent stress sequence of the rotating component;

[0127] Specifically, in this application, the correspondence between the root mean square value of acceleration and the equivalent stress is a predetermined relationship, applicable to analyses of any duration. To verify this hypothesis, the duration of the data segment used to calculate the equivalent stress and root mean square value of acceleration can be gradually compressed until the sample size N approaches 1. Here, the sample can be understood as the number of rotations of the rotating component, and N approaching 1 can be understood as the number of rotations of the rotating component approaching 1 rotation. At this time, the duration of the selected data segment for the equivalent stress of the rotating component coincides with the rotation period of the wheelset system, and there is an expression (17):

[0128]

[0129] Where N represents the rotation frequency of the rotating component, t represents the duration of the data segment for the root mean square value of acceleration, T represents the time taken for the wheelset system to complete one revolution, and σ a This indicates the stress amplitude within the current cycle. It indicates that something leads to, causes, or is deduced from.

[0130] The stress in a rotating component consists of stress amplitude and frequency. The equivalent stress at a frequency of 1 (N→1) is the stress amplitude within that stress cycle. In extreme time intervals (t→T), the stress is determined by… It can be deduced Therefore, it can be concluded that the calculation result of the equivalent stress obtained based on the preset correspondence between the root mean square value of acceleration and the equivalent stress can provide accurate time-domain information prediction.

[0131] Finally, by analyzing the root mean square value of the acceleration of the rotating component, the maximum stress at each measurement point of the rotating component in each rotation cycle can be obtained. When the bearing of the rotating component is subjected to bending moment, the stress on the axial section of the rotating component changes sinusoidally. The specific equivalent stress of the rotating component in one rotation cycle of the wheelset system conforms to expressions (18), (19) and (20):

[0132]

[0133] Where σ represents the equivalent stress of the rotating component at time t. It represents the phase difference, which has the same physical meaning as φ in equation (6) above, but the values ​​are different.

[0134] Step 230: Reconstruct the time domain of the discrete equivalent stress sequence of the rotating component to obtain the time domain information of the stress of the reconstructed rotating component;

[0135] Specifically, time-domain reconstruction of discrete equivalent stress sequences can yield time-domain information on the stress of continuous rotating components.

[0136] The stress assessment method for rotating components provided in this application constructs a stress assessment model for rotating components under the current load environment based on the preset correspondence between the root mean square value of acceleration and equivalent stress, the time-domain signal of the stress spectrum of the rotating component, and the acceleration signal of the rotating component. Then, the model is used to assess the stress state of the rotating component under non-stress sampling time. This application's technical solution successfully establishes the correlation between the acceleration and stress of the rotating component. Through relatively convenient acceleration signal monitoring, it effectively assesses the stress state of the rotating component when it is difficult to monitor. Moreover, the technical solution of this application has universality, providing a convenient, real-time, and non-destructive technical solution for the health monitoring and fault diagnosis of rotating components.

[0137] In a second aspect, this application also provides a stress evaluation device for rotating components, used to implement the above-mentioned stress evaluation method for rotating components, such as... Figure 9 As shown, the device includes:

[0138] The signal acquisition module 310 is configured to acquire the stress spectrum time domain signal and the acceleration signal of the rotating component under the same load environment and within the same sampling time period.

[0139] The fitting module 320 is configured to construct a stress evaluation model of the rotating component under the current load environment based on the preset correspondence between the root mean square value of acceleration and the equivalent stress, the time domain signal of the stress spectrum of the rotating component, and the acceleration signal of the rotating component.

[0140] Evaluation module 330 is configured to evaluate the stress state of the rotating component under non-stress sampling time based on a rotating component stress evaluation model under the current load environment.

[0141] Other preferred embodiments of the rotating component stress evaluation device disclosed in this application achieve the same technical effects as the above-described rotating component stress evaluation method, and will not be repeated here.

[0142] Example:

[0143] In this embodiment, the rotating component is specifically an axle, and the stress state of the axle is tested during train operation.

[0144] The time-domain signal of the axle stress spectrum was obtained using the Smartset testing system. This system consists of two sets of components: the first set is mounted on the wheelset, and the second set is mounted at the end of the equipment compartment. The data sampling frequency for both sets is 1000Hz. Figure 2 As shown, the test sections S3 at the root of the axle unloading groove, S1 on the inner side of the wheel seat, and S2 on the inner side of the gearbox seat were selected as the test sections for this dynamic stress test. The obtained axle stress spectrum is as follows. Figure 3 As shown, the stress spectrum of the axle is processed by rainflow counting (divided into 80 stress levels), and the equivalent stress of the axle section in the corresponding time period is obtained by the above formula (2).

[0145] An axle box vibration acceleration testing system was used to acquire the axle's acceleration signal. An acceleration sensor and data acquisition equipment were connected via a transmission line, with a data sampling frequency of 5000Hz. The axle's acceleration signal was extracted according to the required time period, and this acceleration signal is shown below. Figure 4 As shown, the root mean square value of the axle acceleration within the corresponding time period is calculated using the above formula (3).

[0146] At this point, using 1-hour intervals as the unit, the root mean square of acceleration and the equivalent axle stress were obtained, and a prediction model was established. This model was then used to predict the equivalent stress, such as... Figure 5 As shown, all data fall within the 99% confidence interval. Further shortening the time interval, and combining it with Figure 6, using 20 minutes as the dividing unit, substituting the corresponding root mean square acceleration and axle equivalent stress into the preset correspondence between the root mean square acceleration value and the equivalent stress (Equation 1), the corresponding fitting parameters are obtained as follows: Figure 7 As shown, the operating line represents the carrying environment, R 2 Goodness of fit indicates the quality of the model's fit.

[0147] At this point, for example... Figure 6 As shown, by further reducing the time interval, we can verify whether this rule still applies to the ideal state: when the time interval is small enough to match the rotation period, we can derive the more difficult-to-measure (high time and economic cost) axle dynamic stress signal from the easily measurable acceleration signal using this formula. Furthermore, because when the time matches the rotation period, in the extremely short time it takes for the axle to rotate one revolution, the stress (time-domain stress signal) experienced by the axle can be considered a sinusoidal signal with the equivalent stress as its mean. This allows us to derive the time-domain signal of the axle dynamic stress from the acceleration signal.

[0148] Following the above method, the time is shortened to a division point of one rotation cycle. The obtained fitting parameters and root mean square acceleration are substituted into the calculation formula to obtain the corresponding equivalent stress. Furthermore, theoretically, the time-domain signal of the axle dynamic stress can be obtained based on this equivalent stress result. To verify the reliability of this result, the equivalent stress is also divided into 80 stress levels using the rainflow counting method, and compared with the partial axle stress results obtained from testing. The results are as follows. Figure 8 As shown, the results predicted based on acceleration are in good agreement with the measured results, proving the feasibility of the proposed solution.

[0149] The preferred embodiments of this application have been described in detail above. However, this application is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this application, various simple modifications can be made to the technical solution of this application, and these simple modifications all fall within the protection scope of this application.

[0150] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable way without contradiction. In order to avoid unnecessary repetition, this application will not describe the various possible combinations separately.

[0151] Furthermore, various different embodiments of this application can be combined in any way, as long as they do not violate the spirit of this application, they should also be regarded as the content disclosed by this invention.

Claims

1. A method for stress assessment of rotating components, characterized in that, The method includes: Acquire the time-domain stress spectrum signal and acceleration signal of the rotating component under the same load environment and within the same sampling time period; Based on the preset correspondence between the root mean square value of acceleration and the equivalent stress, the time domain signal of the stress spectrum of the rotating component and the acceleration signal of the rotating component, a stress evaluation model for the rotating component under the current load-bearing environment is constructed. Based on the stress assessment model of the rotating component under the current load-bearing environment, the stress state of the rotating component under non-stress sampling time is evaluated.

2. The method according to claim 1, characterized in that, The expression for the preset root mean square value of acceleration and equivalent stress is as follows: s eq =a RMS k+1 ·10 b ; Where, σ eq a represents the equivalent stress across the cross section of the rotating component. RMS denoted by , where is the root mean square value of the acceleration of the rotating component, and k and b are both fitting parameters of the stress evaluation model for the rotating component.

3. The method according to claim 2, characterized in that, Based on the preset correspondence between the root mean square value of acceleration and the equivalent stress, the time-domain signal of the stress spectrum of the rotating component, and the acceleration signal of the rotating component, a stress assessment model for the rotating component under the current load-bearing environment is constructed, including: Based on the time-domain signal of the stress spectrum of the rotating component, the equivalent stress of the cross-section of the rotating component is calculated, and The root mean square value of the acceleration of the rotating component is calculated based on the acceleration signal of the rotating component. The equivalent stress of multiple sets of rotating component cross sections and the root mean square value of the acceleration of the rotating component are substituted into the preset correspondence between the root mean square value of acceleration and the equivalent stress to obtain the fitting curve; Based on the fitted curve, a stress assessment model for rotating components under the current load-bearing environment is constructed.

4. The method according to claim 3, characterized in that, Based on the time-domain stress spectrum signal of the rotating component, the equivalent stress of the cross section of the rotating component is calculated, including: The stress spectrum time-domain signal of the rotating component is analyzed to obtain the stress amplitude of each stress level in the stress spectrum of the rotating component, the number of cycles at each stress level in the stress spectrum of the rotating component, the total number of cycles in the stress spectrum of the rotating component, and the parameters of the SN curve of the rotating component. The equivalent stress of the rotating component cross section is calculated based on the stress amplitude of each stress level in the stress spectrum of the rotating component, the number of cycles at each stress level in the stress spectrum of the rotating component, the total number of cycles in the stress spectrum of the rotating component, and the parameters of the SN curve of the rotating component.

5. The method according to claim 3, characterized in that, Based on the acceleration signal of the rotating component, the root mean square value of the acceleration of the rotating component is calculated, including: The acceleration signal of the rotating component is analyzed to obtain the acceleration value of each sampling point within the sampling time period and the number of acceleration signals collected within the sampling time period; The root mean square value of the acceleration of the rotating component is calculated based on the acceleration value of each sampling point within the sampling time period and the number of acceleration signals collected within the sampling time period.

6. The method according to claim 1, characterized in that, After constructing a stress assessment model for rotating components under the current load-bearing environment, the method further includes: Verify whether the goodness of fit of the stress assessment model for rotating components under the current load-bearing environment meets the preset requirements; If satisfied, the stress state of the rotating component under non-stress sampling time is evaluated according to the stress evaluation model of the rotating component under the current load environment. If the requirements are not met, the fitting parameters are reselected until the goodness of fit of the stress evaluation model of the rotating component meets the preset requirements.

7. The method according to claim 1, characterized in that, Based on the stress assessment model for rotating components under the current load-bearing environment, the stress state of the rotating component under non-stress sampling time is assessed, including: Acquire the acceleration signal of the rotating component during the non-stress sampling time; Based on the acceleration signal of the rotating component during the non-stress sampling time and the stress evaluation model of the rotating component, the time-domain information of the corresponding rotating component stress is obtained; Based on the time-domain information of the stress of the rotating component, the stress state of the rotating component under non-stress sampling time is determined.

8. The method according to claim 7, characterized in that, Based on the acceleration signal of the rotating component during the non-stress sampling time and the stress evaluation model of the rotating component, the time-domain information of the corresponding rotating component stress is obtained, including: The acceleration signal of the rotating component under non-stress sampling time is divided according to the rotation period of the wheelset system under the current load environment to obtain multiple segments of acceleration signal under non-stress sampling time. The acceleration signals under the multiple non-stress sampling times are fed into the stress evaluation model of the rotating component to obtain a discrete equivalent stress sequence of the rotating component; The time-domain information of the stress of the discrete rotating component is obtained by reconstructing the equivalent stress sequence in the time domain.

9. A stress evaluation device for rotating components, characterized in that, The device includes: The signal acquisition module is configured to acquire the stress spectrum time-domain signal and acceleration signal of the rotating component under the same load environment and within the same sampling time period; The fitting module is configured to construct a stress evaluation model of the rotating component under the current load environment based on the preset correspondence between the root mean square value of acceleration and the equivalent stress, the time domain signal of the stress spectrum of the rotating component, and the acceleration signal of the rotating component. The evaluation module is configured to evaluate the stress state of the rotating component under non-stress sampling time based on the rotating component stress evaluation model under the current load environment.

10. The apparatus according to claim 9, characterized in that, The expression for the preset root mean square value of acceleration and equivalent stress is as follows: s eq =a RMS k+1 ·10 b ; Where, σ eq a represents the equivalent stress across the cross section of the rotating component. RMS denoted by , where is the root mean square value of the acceleration of the rotating component, and k and b are both fitting parameters of the stress evaluation model for the rotating component.

Images