Accelerated impact fatigue life prediction method considering load uncertainty

Abstract

The invention discloses an accelerated impact fatigue life prediction method considering load uncertainty, and belongs to the field of accelerated impact fatigue life of spacecraft electronic components. Comprising the following steps: S1, selecting impact response spectrum SRS parameters and carrying out distribution modeling; s2, constructing an impact fatigue life database; s3, carrying out Weibull distribution fitting on the fatigue life; s4, identifying the function relationship between the Weibull parameter and the SRS to quantify the load dispersion effect; s5, analyzing a Weibull parameter change rule; s6, establishing an accelerated impact fatigue life model; s7, predicting an impact fatigue damage boundary; and S8, model verification. Aiming at the non-stationary and non-Gaussian characteristics of the complex impact load, a special accelerated fatigue life model is established, and the technical blank in the field is filled. By analyzing the change rule of the Weibull distribution shape parameters, the constants and the abnormal numbers are distinguished, the corresponding correction model is established, and the applicability and the accuracy of the model are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to an accelerated impact fatigue life prediction method considering load uncertainty, belonging to the field of accelerated impact fatigue life prediction of spacecraft electronic components. BACKGROUND

[0002] With the development of reusable spacecraft technology, spacecraft electronic components need to withstand multiple complex impact loads, including booster ignition, explosive device explosion and landing impact, etc. These impact loads have the characteristics of short duration, large load amplitude and wide frequency range, which are more complex than traditional simplified half-sine drop impact.

[0003] In aerospace engineering, when the load amplitude is high, the corresponding fatigue life is relatively short, the test duration is short, and the cost is controllable. However, at a lower load amplitude, the fatigue life increases significantly, resulting in an excessively long test duration. Considering the time and budget limitations of aerospace engineering, high-life fatigue testing is often impractical.

[0004] In the prior art, the research on accelerated fatigue life model is mainly based on random vibration fatigue, mainly aiming at stationary random vibration. However, structures subjected to complex impact loads usually exhibit non-stationary and non-Gaussian random vibration characteristics. Due to the randomness of impact loads and the characteristics of impact response spectrum, the same impact response spectrum may correspond to multiple time-domain loads, resulting in time-domain load uncertainty and further leading to the dispersion of impact fatigue life.

[0005] The prior art has the following problems:

[0006] 1. Model limitations: Existing acceleration models (such as Allegri inverse power law model) are only applicable to stationary Gaussian random vibration, while spacecraft complex impact loads (such as booster ignition, explosive impact) have non-stationary and non-Gaussian characteristics, resulting in the failure of traditional PSD (power spectral density) acceleration model methods.

[0007] 2. Ignoring load uncertainty: There are multiple time-domain load phase combinations within the same ±6dB envelope of the impact response spectrum (SRS). The existing method does not quantify the influence of dispersion caused by load uncertainty on life distribution.

[0008] 3. Time-consuming and high-cost testing: In the development process of reusable spacecraft, impact durability testing of electronic component welds is essential. However, these tests involve an iterative trial-and-error process of continuous design, manufacturing and testing until mission requirements are met. This process not only consumes a large amount of resources, but also greatly prolongs the development schedule of the spacecraft. SUMMARY

[0009] The purpose of the present invention is to solve the above-mentioned problems existing in the background technology and to provide a method for predicting accelerated impact fatigue life taking into account load uncertainty.

[0010] The present invention achieves the above-mentioned purpose by adopting the following technical solutions:

[0011] A method for predicting accelerated impact fatigue life considering load uncertainty, the method comprising the following steps:

[0012] S1: Shock response spectrum (SRS) parameter selection and distribution modeling;

[0013] S2: Construct impact fatigue life database;

[0014] S3: Weibull distribution fitting of fatigue life;

[0015] S4: Identify the functional relationship between Weibull parameter and SRS to quantify the load dispersion effect;

[0016] S5: Analysis of the variation pattern of Weibull parameters;

[0017] S6: Establishment of accelerated impact fatigue life model;

[0018] S7: Impact fatigue damage boundary prediction;

[0019] S8: Model validation.

[0020] Compared with the prior art, the present invention has the following beneficial effects: 1. Fully consider the load dispersion effect: This invention systematically considers the influence of impact response spectrum load dispersion on fatigue life for the first time, establishes a quantitative relationship between load dispersion and fatigue life dispersion, and improves the prediction accuracy.

[0021] 2. Applicable to complex impact loads: Aiming at the non-stationary and non-Gaussian characteristics of complex impact loads, a special accelerated fatigue life model was established, filling the technical gap in this field.

[0022] 3. Considering the non-constant characteristics of Weibull parameters: By analyzing the changing pattern of the shape parameters of the Weibull distribution, the constant and non-constant cases are distinguished, and the corresponding correction model is established to improve the applicability and accuracy of the model.

[0023] 4. High prediction accuracy: Experimental verification shows that the relative error between the predicted value and the actual value is in the range of -5%~7%, which meets the requirements of engineering applications.

[0024] 5. Significantly improve test efficiency: Predict low-load, long-life performance through high-load, short-life tests, significantly shortening test time and reducing test costs.

[0025] 6. Strong engineering practicality: It provides a complete technical solution from database establishment to model application, which has strong engineering practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is an overall flow chart of an accelerated impact fatigue life prediction method considering load uncertainty of the present invention;

[0027] Figure 2 It is a schematic diagram of shock response spectrum parameters of an accelerated shock fatigue life prediction method considering load uncertainty of the present invention;

[0028] Figure 3 It is the Weibull scale parameter of the accelerated impact fatigue life prediction method considering load uncertainty of the present invention. At different inflection point amplitudes Logarithmic fitting relationship plot and shape parameter under At different inflection point amplitudes The linear fitting graph below;

[0029] Figure 4 The present invention is a method for predicting the fatigue life of an accelerated impact considering load uncertainty. and Fitting coefficients ( 、 、 、 ) at different inflection point frequencies and low frequency slope Three-dimensional surface distribution diagram under the combination;

[0030] Figure 5 The present invention is a method for predicting the life of an accelerated impact fatigue life considering load uncertainty, which is a three-dimensional surface diagram of the impact fatigue failure boundary prediction result, wherein the X axis is , the Y axis is , the Z axis is the corresponding failure acceleration boundary ;

[0031] Figure 6 This is an error analysis diagram of the model prediction value and the actual accelerated test result of the accelerated impact fatigue life prediction method considering load uncertainty of the present invention. DETAILED DESCRIPTION

[0032] The technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0033] Specific implementation method 1: Figures 1-6 As shown, this embodiment describes a method for predicting the life of accelerated impact fatigue considering load uncertainty, taking the electronic packaging of BGA solder joints as an example, referring to Figure 1 , including the following steps:

[0034] S1: Construct a shock response spectrum (SRS) parameter set and generate n (n=40 in this example) time-domain shock loads for each SRS;

[0035] In S1, the SRS parameter set under a typical aerospace shock environment is selected, including:

[0036] SRS inflection point amplitude ( ): 2000g to 5000g; low frequency slope ( ): 6~14 dB / Oct; inflection point frequency ( ): 500Hz to 2000Hz;

[0037] like Figure 2 As shown in the figure, the spectrum of SRS is intuitively shown, in which the 、 and By setting a ±6dB tolerance range through the test standard, a series of time domain loads with different dispersion degrees were constructed, providing a basis for subsequent load dispersion modeling;

[0038] S2: Constructing an impact fatigue life database

[0039] The wavelet transform method is used to perform time domain inversion on the target SRS to generate multiple groups of impact load signals that are within the tolerance range of ±6dB of the SRS but have different time domain waveforms and phases.

[0040] A three-dimensional finite element model of the target structure was established and verified experimentally. The synthesized time-domain impact loads were applied as input to the boundary nodes of the finite element model to simulate the stress response of the BGA solder joints under various load groups.

[0041] Extract the stress-time history of the maximum stress point at the bottom of the BGA solder joint and use Formula 1 and Formula 2 to calculate the single impact damage respectively , and impact fatigue life ;

[0042] (1)

[0043] (2)

[0044] Is to judge Is it a function of a broadband or narrowband random process? If it is a broadband random process, then is calculated; if it is a narrowband random process, then is calculated; T is the total duration of the non-stationary process under a single shock; is the spectral moment; and They are kurtosis and skewness of the segment; and It's the material Parameters of the curve;

[0045] By conducting multiple sets of load tests and finite element calculations, we obtain life data sets, providing sufficient samples for subsequent statistical analysis;

[0046] S3: Weibull distribution fitting for fatigue life

[0047] The lifespan database obtained by S2 was statistically analyzed using the Weibull distribution model according to Formula 3, and the Weibull parameters were solved by combining Bootstrap and Bayesian estimation.

[0048] (3)

[0049] in is the scale parameter, is the shape parameter;

[0050] The structural reliability function can be expressed by formula 4 as

[0051] (4)

[0052] Therefore, reliability Fatigue life under It can be expressed by formula 5 as

[0053] (5)

[0054] Bootstrap and Bayesian algorithms are used to estimate the Weibull distribution parameters under different SRS parameters. Figure 3 shown.

[0055] S4: Identify the functional relationship between Weibull parameter and SRS to quantify the load dispersion effect;

[0056] By S3 and Figure 3 The relationship between the obtained Weibull parameter and SRS can be found that when the SRS and When it remains unchanged, It decreases monotonically with the increase of the SRS inflection point amplitude. The logarithm of and the logarithm of the SRS inflection point amplitude show a strong linear relationship. Under the condition that other SRS parameters are constant, It increases in proportion to the amplitude of the SRS inflection point and shows a linear correlation.

[0057] Establish and The logarithmic linear relationship is expressed as follows:

[0058] (6)

[0059] Establish and The linear relationship is expressed as follows:

[0060] (7)

[0061] S5: Analysis of the variation pattern of Weibull parameters

[0062] In S4, the slope of the fitting curve 、 and intercept 、 Follow and The relationship needs to be obtained by data surface fitting. Figure 4 As shown in Table 1, the corresponding mathematical model can be obtained through polynomial regression, which quantifies the influence of load dispersion effect on fatigue life distribution.

[0063]

[0064] Table 1

[0065] Depend on Figure 4 The parameters are obtained by the mathematical model quantified in Table 1 ( 、 、 、 ) with the distribution of SRS parameters, where >1500Hz and >8 o'clock, The change range of is small, indicating that it can be approximated as a constant; under other conditions, This analysis provides a basis for the subsequent accelerated impact fatigue life model. It provides a basis for distinguishing between constant constants and non-constant constants.

[0066] S6: Accelerated impact fatigue life model establishment

[0067] Established according to S4 and Logarithmic linear relationship, impact fatigue life distribution before the impact load increases It can be expressed by formula 8:

[0068] (8)

[0069] Impact fatigue life distribution with increasing impact load magnitude It can be expressed by formula 9:

[0070] (9)

[0071] From formula 9 to formula 8, we can get formula 10, the inverse power law random accelerated fatigue life model

[0072] (10)

[0073] when When is a constant, the reliability after the load level increases is equal to the reliability before the load level increases, then the corresponding impact fatigue life relationship is

[0074] (11)

[0075] when When it is not a constant, Follow changes, the load distribution correction factor is introduced , establish a correction model

[0076] (12)

[0077] in It can be expressed by formula 13

[0078] (13)

[0079] By analyzing the parameters of SRS in different and Under the condition of surface fitting, the >1500Hz and >8 o'clock, can be approximated as a constant; under other conditions, It behaves as a non-constant constant. Determine the constant and non-constant The accelerated impact fatigue life model under the above conditions is Equation 14:

[0080] (14)

[0081] S7: Impact Fatigue Damage Boundary Prediction

[0082] Based on the accelerated life model established in step S6, combined with the target design life and specified reliability , the fatigue failure boundary under the original working condition is reversely calculated and defined as the impact fatigue damage boundary;

[0083] like Figure 5 As shown, and The horizontal and vertical axes are the damage failure boundary value and the Z axis constitutes a three-dimensional surface, which can intuitively reflect the safety margin under different load distribution conditions. and Parameters, their corresponding values ​​are estimated by interpolation method to achieve comprehensive evaluation;

[0084] S8: Model verification and parameter correction

[0085] In order to verify the accuracy of the above accelerated life model, actual accelerated impact fatigue tests were carried out under multiple SRS parameter conditions, and the life of each sample was recorded. .

[0086] like Figure 6 As shown in the figure, the relative error between the model-predicted life and the actual test life is compared. The error of all samples is controlled within ±7%.

[0087] Based on the verification results, the fitting parameters or correction items can be further adjusted to ensure the prediction accuracy and reliability of the model in engineering applications.

[0088] 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 invention can be implemented in other configurations without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations coming within the meaning and range of equivalents of the claims are intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

[0089] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A method for predicting accelerated impact fatigue life considering load uncertainty, characterized by: The method comprises the following steps: S1: Shock response spectrum (SRS) parameter selection and distribution modeling; S2: Construct impact fatigue life database; S3: Weibull distribution fitting of fatigue life; S4: Identify the functional relationship between Weibull parameter and SRS to quantify the load dispersion effect; S5: Analysis of the variation pattern of Weibull parameters; S6: Establishment of accelerated impact fatigue life model; S7: Impact fatigue damage boundary prediction; S8: Model validation.

2. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 1, characterized in that: The SRS parameters in S1 include the low frequency slope , inflection point amplitude , inflection point frequency , used to build a foundation for load distribution modeling.

3. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 2, characterized in that: The SRS inflection point amplitude Value range: 2000g to 5000g; low frequency slope Value range: 6 to 14 ; Inflection point frequency Value range: 500Hz to 2000Hz.

4. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 1 or 3, characterized in that: Said S2 further comprises: S2.1: A time-domain load synthesis method based on wavelet analysis is used to generate multiple corresponding time-domain load signals for each SRS parameter combination; S2.2: Use finite element analysis to calculate the impact response of key failure locations of the structure. Combine finite element simulation with fatigue damage model to establish a fatigue life dataset for each set of SRS parameters corresponding to time domain loads.

5. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 4, characterized in that: The S3 method combines Bootstrap and Bayesian estimation to fit the Weibull distribution parameters of life data. and , is the scale parameter, is the shape parameter, and the corresponding relationship between it and the SRS parameter is obtained.

6. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 5, characterized in that: Said S4 further comprises: S4.1: Establishing the scale parameter of the Weibull distribution The logarithm of the shock response spectrum inflection point amplitude Linear relationship between logarithms: S4.2: Establishing the Weibull Distribution Shape Parameters and the amplitude of the inflection point of the shock response spectrum The linear relationship between: in, 、 、 、 are the corresponding slope and intercept parameters, respectively.

7. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 6, characterized in that: The S5 further includes: S5.1: Analyzing Parameters by Data Fitting 、 、 、 Shock response spectrum parameters and the law of change; S5.2: Determine Weibull Shape Parameters is the range of shock response spectrum parameters that are constant and non-constant.

8. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 7, characterized in that: The S6 further includes: S6.1: For When is a constant, a simplified inverse power law model is adopted: S6.2: For For non-constant cases, a relative dispersion factor is introduced , establish a modified accelerated fatigue life model: in, is the test life after acceleration, is the original test life, is the correction factor: For fatigue life reliability when S6.3: Considering the load dispersion effect by combining S6.1 and S6.2, the final accelerated impact fatigue life model is: in, 、 、 、 Represent the inflection point frequencies of the shock response spectrum and low frequency slope The values ​​within the range of each stage of the model.

9. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 8, characterized in that: The S7 further includes: S7.1: Calculate the given reliability based on the established accelerated fatigue life model and impact fatigue damage boundaries under design life; S7.2: Construct and The horizontal and vertical coordinates are the three-dimensional surface with the impact fatigue damage boundary value as the Z axis; S7.3: For those not covered in the trial and Conditions are calculated and the corresponding values ​​are estimated using interpolation methods.

10. The method for predicting accelerated impact fatigue life considering load uncertainty according to claim 1, characterized in that: The S8 further includes: S8.1: Verify the model prediction accuracy through actual impact fatigue tests; S8.2: Calculate the relative error between the predicted value and the actual value; S8.3: Adjust model parameters based on validation results to ensure that prediction errors are within acceptable limits.

Images