Vibration fatigue life prediction method considering random factor influence, electronic equipment and storage medium

Abstract

The invention discloses a vibration fatigue life prediction method considering the influence of random factors, electronic equipment and a storage medium, and the method comprises the steps: carrying out the calculation of a power spectrum of a random vibration response time domain signal, obtaining a random vibration response spectrum moment, and judging whether a vibration response type belongs to narrowband response or broadband response; establishing a life prediction model corresponding to the vibration response type; fitting distribution parameters of corresponding distribution are obtained through a statistical method, fitting distribution is obtained, and goodness-of-fit test is conducted on the fitting distribution; performing Monte Carlo sampling on the fitting distribution passing the goodness-of-fit test to obtain a sampling result; and inputting a sampling result into the life prediction model to obtain a fatigue life sample, and obtaining a fatigue life empirical distribution function after the fatigue life sample passes a goodness test. According to the failure life distribution prediction method provided by the invention, the probability distribution of the fatigue life is obtained, the confidence coefficient of a fatigue analysis result is improved, and engineering decision reference is facilitated.

Description

Technical Field

[0001] The invention relates to the field of vibration fatigue analysis, and in particular to a vibration fatigue life prediction method, electronic equipment and storage medium taking into account the influence of random factors. Background Art

[0002] Aerospace structures, ship and marine facilities, and railway and road vehicles are in a random vibration environment for a long time. Their connection structures and working parts are susceptible to random vibration, which leads to vibration fatigue failure, causing failure of important connection structures of products, damage to working parts, and other failures. According to the fatigue characteristics of materials, the cyclic load that causes fatigue failure is much smaller than the tensile strength limit under static load, which has a very high risk of vibration fatigue damage and is very likely to cause economic losses and social harm. Therefore, it is of great research value to analyze and predict the vibration fatigue life of structures.

[0003] At present, there are two main methods for analyzing vibration fatigue life: time domain analysis and frequency domain analysis. Although the time domain analysis method based on rain flow counting is intuitive, accurate and widely used, it needs to process a large amount of time series data, which wastes time and computing resources. In comparison, the frequency domain method of calculating fatigue damage through the response power spectrum of the structure is more feasible and efficient. However, in actual engineering, due to the influence of various random factors, such as material microstructure, defect content, processing technology, residual stress and load level, fatigue cumulative damage has inherent dispersion. At the same time, measurement errors that cannot be eliminated will be introduced when collecting data. When the existing frequency domain method is used to calculate the vibration fatigue life, the randomness of fatigue damage and the dispersion of estimated values ​​are not taken into account, and the calculation results cannot reflect the statistical characteristics of fatigue life. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide a vibration fatigue life prediction method, electronic equipment and storage medium taking into account the influence of random factors.

[0005] The objective of the present invention is achieved through the following technical solutions:

[0006] A first aspect of the present invention provides a vibration fatigue life prediction method considering the influence of random factors, comprising:

[0007] S1: Calculate the power spectrum of the random vibration response time domain signal to obtain the random vibration response spectrum moment, and further use the random vibration response spectrum moment to perform bandwidth analysis to determine whether the vibration response type belongs to narrowband response or broadband response;

[0008] S2: Establish a life prediction model corresponding to the vibration response type;

[0009] S3: treating the random vibration response spectrum moment as a group of random samples for statistical analysis, assuming that it obeys the corresponding distribution, using statistical methods to obtain the fitting distribution parameters of the corresponding distribution and obtain the fitting distribution, and performing a goodness of fit test on the fitting distribution;

[0010] S4: Perform Monte Carlo sampling on the fitting distribution that passes the goodness of fit test to obtain the sampling results;

[0011] S5: Input the sampling results into the life prediction model to obtain fatigue life samples, and obtain fatigue life empirical distribution function after the fatigue life samples pass the goodness of fit test.

[0012] Furthermore, in step S1, the power spectrum of the random vibration response time domain signal is calculated to obtain the random vibration response spectral moment, including:

[0013] S101: Based on the density of the power spectrum of the random vibration response time domain signal, the i-th order spectral moment m of the random process is obtained. i for:

[0014]

[0015] Where G(f) is the power spectrum density; f is the frequency;

[0016] In step S1, the use of random vibration response spectrum moment to perform bandwidth analysis to determine whether the vibration response type belongs to narrowband response or broadband response includes:

[0017] S102: Calculate the 0th order spectral moment m0, the 2nd order spectral moment m2, and the 4th order spectral moment m4 respectively;

[0018] S103: Calculate the irregularity factor γ of the random process using the 0th order spectral moment m0, the 2nd order spectral moment m2, and the 4th order spectral moment m4:

[0019]

[0020] S104: Calculate the bandwidth coefficient ε describing the bandwidth of the random process using the irregular factor γ:

[0021]

[0022] When the bandwidth coefficient ε approaches 0, the random process conforms to a narrowband random process, and the vibration response type is judged to be a narrowband response; when the bandwidth coefficient ε approaches 1, the random process corresponds to a broadband random process, and the vibration response type is judged to be a broadband response.

[0023] Furthermore, in step S2, the establishing of the life prediction model corresponding to the vibration response type includes establishing a narrowband response life prediction model and establishing a broadband response life prediction model.

[0024] Furthermore, the establishment of the narrowband response life prediction model is to establish a narrowband response structure life prediction model based on a narrowband approximation method, which specifically includes:

[0025] S201: Based on the SN curve and the linear cumulative damage Miner criterion, the damage amount of the structure within time T is obtained, including:

[0026] S20101: Based on the SN curve, fatigue life is predicted, and the stress amplitude σ and fatigue life N σ The following first relationship exists:

[0027] σ b N σ =A

[0028] In the formula, b and A are constants, which are obtained by fitting the material fatigue performance test data;

[0029] S20102: Based on the linear cumulative damage Miner criterion, the structural damage D of the structure within the action time T is:

[0030]

[0031] In the formula, The amplitude is σ i The number of times the stress acts at time T, is the stress amplitude σ i Fatigue life under, N is stress σ i The average number of cycles within the range of Δσ, p(σ i ) represents the stress amplitude σ i The probability density function of

[0032] The average number of cycles in the stress range is obtained by the following formula:

[0033] N=N0T

[0034] Where N0 is the stress per unit time σ i The average number of cycles within the range Δσ;

[0035] S20103: Substitute the first relationship in step S20101 into the damage amount formula in step S20102 to obtain the structural damage amount D:

[0036]

[0037] When the response is a narrowband vibration, N0 = E(p), where E(p) is the expectation of the peak frequency of the random response signal. The peak frequency of the narrowband process is close to the zero crossing frequency, and the expectation is expressed as:

[0038]

[0039] In the formula, ω n is the structural natural frequency;

[0040] S202: Using the one-to-one correspondence between the peaks and troughs of the narrowband random process, the stress amplitude p(σ) obeying the Rayleigh distribution is obtained as the probability density function of the stress amplitude:

[0041]

[0042] In the formula, σ RMS is the root mean square value of the structural dynamic response force,

[0043] S203: Substitute the stress amplitude p(σ) calculated in step S202 as the probability density function of the stress amplitude into the expression of the structural damage amount D calculated in step S201 based on the linear cumulative damage Miner criterion, and assume that the constant amplitude SN curve has a linear characteristic on the double logarithmic graph, and obtain the damage accumulation D under the narrow band action NB (T):

[0044]

[0045] S204:D NB When (T)=1, fatigue failure occurs, and the fatigue life, i.e., the narrow-band response structure life prediction model, is:

[0046]

[0047] Furthermore, the establishment of a broadband response life prediction model is specifically based on α 0.75 Methods A broadband response structure life prediction model is established, including:

[0048] S205: Introducing parameter α 0.75 , correct the damage accumulation D under narrowband action calculated in step S203 NB , we can get the damage accumulation D under broadband action WB :

[0049]

[0050] The parameter α 0.75 Defined as:

[0051]

[0052] In the formula, m0, m 0.75 and m 1.5 They are the 0th order spectral moment, 0.75th order spectral moment and 1.5th order spectral moment of the random process respectively;

[0053] S206:D WB When (T) = 1, fatigue failure occurs, and the fatigue life is the broadband response structure life prediction model:

[0054]

[0055] Among them, the parameter β is:

[0056]

[0057] Further, in step S3, for the narrowband random process, the random vibration response spectrum moment is regarded as a group of random samples for statistical analysis, assuming that it obeys the corresponding distribution, a statistical method is used to obtain the fitting distribution parameters of the corresponding distribution and obtain the fitting distribution, and a goodness of fit test is performed on the fitting distribution, including:

[0058] S301: For a narrowband random process, assuming σ RMS is a random variable and follows a normal distribution, that is, σ RMS ~N(μ gsg ,σ gsg ), we get σ RMS Distribution parameter μ gsg and σ gsg for;

[0059] S302: Use the Kolmogorov-Smirnov method to calculate σ RMS ~N(μ gsg ,σ gsg ) to conduct the test, and accept the null hypothesis when the significance level is 5%, that is, σ RMS Satisfy σ RMS ~N(μ gsg ,σ gsg );

[0060] In step S4, for the narrow-band random process, Monte Carlo sampling is performed on the fitting distribution that passes the goodness of fit test to obtain a sampling result, including:

[0061] S401: Based on the σ obtained in step S302 RMS Fitting distribution σ RMS ~N(μ gsg ,σ gsg ) for σ RMS Perform n samplings and obtain n sampling results.

[0062] Further, in step S3, for the broadband random process, the random vibration response spectrum moment is regarded as a group of random samples for statistical analysis, assuming that it obeys the corresponding distribution, a statistical method is used to obtain the fitting distribution parameters of the corresponding distribution and obtain the fitting distribution, and a goodness of fit test is performed on the fitting distribution, including:

[0063] S303: For a broadband random process, according to the parameter β of the random process, similarly, assuming that β obeys a normal distribution β~N(μ β ,σ β ), and obtain the β distribution parameter μ β and σ β ;

[0064] S304: Use the Kolmogorov-Smirnov method to calculate β~N(μ β ,σ β ) to conduct the test, and accept the null hypothesis when the significance level is 5%, that is, β~N(μ β ,σ β );

[0065] In step S4, for the broadband random process, Monte Carlo sampling is performed on the fitting distribution that passes the goodness of fit test to obtain a sampling result, including:

[0066] S401: Fitting distribution β~N(μ β ,σ β ) Sample β n times and obtain n sampling results.

[0067] Furthermore, in step S5, the sampling result is input into the life prediction model to obtain a fatigue life sample, and after the fatigue life sample passes the goodness test, a fatigue life empirical distribution function is obtained, including:

[0068] S501: inputting the n sampling results obtained in step S4 into the life prediction model obtained in step S2 to obtain n fatigue life samples;

[0069] S502: using the Kolmogorov-Smirnov goodness of fit test, the n fatigue life samples obtained in step S501 are sequentially tested for log-normal distribution, Weibull distribution and exponential distribution hypothesis, and when a 5% significance level is obtained, it is considered that the life of the narrow-band / broadband vibration response product obeys the log-normal distribution;

[0070] S503: Plotting fatigue life empirical distribution functions corresponding to the n sampling results obtained in step S4.

[0071] According to a second aspect of the present invention, there is provided an electronic device, comprising a storage unit and a processing unit, wherein the storage unit stores computer instructions executable on the processing unit, and the processing unit executes the steps of the method described in the first aspect when executing the computer instructions.

[0072] According to a third aspect of the present invention, there is provided a storage medium having computer instructions stored thereon, wherein the computer instructions, when executed, execute the steps of the method described in the first aspect.

[0073] The beneficial effects of the present invention are:

[0074] In an exemplary embodiment of the present invention, different life prediction models are applied to narrowband and broadband signals, and the randomness of fatigue damage is considered. The fatigue life is predicted by combining the Monte Carlo method to obtain the predicted probability distribution of fatigue life. Fatigue analysis is carried out based on material fatigue test data and structural vibration response, without the need for time domain load spectrum, which simplifies the requirements for fatigue analysis input conditions and improves analysis efficiency. A failure life distribution prediction method is also given, and the probability distribution of fatigue life is obtained, which improves the confidence of fatigue analysis results and is more conducive to engineering decision-making reference. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A flowchart of a vibration fatigue life prediction method considering the influence of random factors provided in an exemplary embodiment of the present invention;

[0076] Figure 2 is the RMS value σ of the narrowband power spectrum density provided in an exemplary embodiment of the present invention RMS Schematic diagram of sampling results;

[0077] Figure 3 A schematic diagram of a sampling result of a broadband power spectral density eigenvalue β provided in an exemplary embodiment of the present invention;

[0078] Figure 4 A schematic diagram of fatigue life corresponding to a sampling result of a narrow-band power spectrum density root mean square value provided in an exemplary embodiment of the present invention;

[0079] Figure 5 A schematic diagram of a life prediction result of a narrow-band vibration response product provided in an exemplary embodiment of the present invention;

[0080] Figure 6 A schematic diagram of fatigue life corresponding to a sampling result of a broadband power spectrum density characteristic value provided in an exemplary embodiment of the present invention;

[0081] Figure 7 It is a schematic diagram of the life prediction results of a broadband vibration response product provided in an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0082] The technical solution of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0083] In the description of the present invention, it should be noted that the directions or positional relationships indicated by "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. are directions or positional relationships based on the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0084] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, "installation", "connection" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0085] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0086] See also Figure 1 , Figure 1 A flow chart of a vibration fatigue life prediction method considering the influence of random factors provided in an exemplary embodiment of the present invention is shown, comprising:

[0087] S1: Calculate the power spectrum of the random vibration response time domain signal to obtain the random vibration response spectrum moment, and further use the random vibration response spectrum moment to perform bandwidth analysis to determine whether the vibration response type belongs to narrowband response or broadband response;

[0088] S2: Establish a life prediction model corresponding to the vibration response type;

[0089] S3: treating the random vibration response spectrum moment as a group of random samples for statistical analysis, assuming that it obeys the corresponding distribution, using statistical methods to obtain the fitting distribution parameters of the corresponding distribution and obtain the fitting distribution, and performing a goodness of fit test on the fitting distribution;

[0090] S4: Perform Monte Carlo sampling on the fitting distribution that passes the goodness of fit test to obtain the sampling results;

[0091] S5: Input the sampling results into the life prediction model to obtain fatigue life samples, and obtain fatigue life empirical distribution function after the fatigue life samples pass the goodness of fit test.

[0092] Specifically, in this exemplary embodiment, a vibration fatigue life prediction method considering the influence of random factors is proposed based on the frequency domain method, taking into account the randomness of fatigue damage and the dispersion of estimated values: first, a power spectrum analysis is performed on the time-domain vibration response of the structure to determine whether the response is a narrowband / broadband random process (step S1); then, a spectral moment-life model is established for narrowband and broadband random processes, respectively (step S2), and the response spectral moment is probabilistically fitted (step S3); finally, the Monte Carlo method is used to randomly sample the spectral moment (step S4), and the sample fatigue life is calculated and probabilistically fitted based on the established spectral moment-life model to obtain the fatigue life distribution characteristics of the structure (step S5).

[0093] Therefore, this exemplary embodiment aims at the problem that the traditional deterministic vibration fatigue analysis method cannot consider the structural durability difference and the random phenomenon in the fatigue physical process, reduces the deviation of fatigue life prediction, and solves the problem of low reliability of engineering application of the traditional fatigue life prediction method:

[0094] Different life prediction models are applied to narrowband and broadband signals, and the randomness of fatigue damage is considered. The Monte Carlo method is combined to predict fatigue life and obtain the predicted probability distribution of fatigue life. Fatigue analysis is carried out based on material fatigue test data and structural vibration response, without the need for time domain load spectrum, which simplifies the requirements for fatigue analysis input conditions and improves analysis efficiency. A failure life distribution prediction method is also given, and the probability distribution of fatigue life is obtained, which improves the confidence of fatigue analysis results and is more conducive to engineering decision-making reference.

[0095] The following content will describe in detail the preferred exemplary embodiments of each step, and use Matlab to simulate 10 narrowband and broadband time-domain random vibration response signal samples respectively:

[0096] More preferably, in an exemplary embodiment, in step S1, calculating the power spectrum of the random vibration response time domain signal to obtain the random vibration response spectral moment includes:

[0097] S101: Based on the density of the power spectrum of the random vibration response time domain signal, the i-th order spectral moment m of the random process is obtained. i for:

[0098]

[0099] Where G(f) is the power spectrum density; f is the frequency;

[0100] Among them, the power spectrum density root mean square value σ RMS Often used to describe the strength of random vibration, σ RMS The larger the value, the stronger the vibration. RMS The relationship with the 0th order spectral moment is:

[0101]

[0102] This will be covered in the next steps.

[0103] In step S1, the use of random vibration response spectrum moment to perform bandwidth analysis to determine whether the vibration response type belongs to narrowband response or broadband response includes:

[0104] S102: Calculate the 0th order spectral moment m0, the 2nd order spectral moment m2, and the 4th order spectral moment m4 respectively;

[0105] S103: Calculate the irregularity factor γ of the random process using the 0th order spectral moment m0, the 2nd order spectral moment m2, and the 4th order spectral moment m4:

[0106]

[0107] S104: Calculate the bandwidth coefficient ε describing the bandwidth of the random process using the irregular factor γ:

[0108]

[0109] When the bandwidth coefficient ε approaches 0, the random process conforms to a narrowband random process, and the vibration response type is judged to be a narrowband response; when the bandwidth coefficient ε approaches 1, the random process corresponds to a broadband random process, and the vibration response type is judged to be a broadband response.

[0110] In specific applications, the Shang formula is used to verify the bandwidths of the two signals, and the bandwidth coefficient ε1 of the narrowband signal is obtained to be about 0.13, and the bandwidth coefficient ε2 of the broadband signal is obtained to be about 0.54, which meets the definition of bandwidth coefficient.

[0111] More preferably, in an exemplary embodiment, in step S2, the establishing of the life prediction model corresponding to the vibration response type includes establishing a narrowband response life prediction model and establishing a broadband response life prediction model.

[0112] Specifically, based on the spectral moment method, narrowband response and broadband response structural life prediction models are established respectively. After the power spectrum analysis of the structural response of a certain number of products under the excitation load, the spectral moment is obtained and the stress amplitude probability density function (PDF) is calculated. Combined with the fatigue damage accumulation theory and the SN curve prediction of the material, the structural fatigue life prediction is realized. The vibration fatigue life prediction model is established for the amplitude probability density function (PDF) of the narrowband process vibration response and the broadband process vibration response.

[0113] More preferably, in an exemplary embodiment, the establishing of the narrowband response life prediction model is to establish a narrowband response structure life prediction model based on a narrowband approximation method, specifically including:

[0114] S201: Based on the SN curve and the linear cumulative damage Miner criterion, the damage amount of the structure within time T is obtained, including:

[0115] S20101: Based on the SN curve, fatigue life is predicted, and the stress amplitude σ and fatigue life N σ The following first relationship exists:

[0116] σ b N σ =A

[0117] In the formula, b and A are constants, which are obtained by fitting the material fatigue performance test data;

[0118] S20102: Based on the linear cumulative damage Miner criterion, the structural damage D of the structure within the action time T is:

[0119]

[0120] In the formula, The amplitude is σ i The number of times the stress acts at time T, is the stress amplitude σ i Fatigue life under, N is stress σ i The average number of cycles within the range of Δσ, p(σ i ) represents the stress amplitude σ i The probability density function of

[0121] The average number of cycles in the stress range is obtained by the following formula:

[0122] N=N0T

[0123] Where N0 is the stress per unit time σ i The average number of cycles within the range Δσ;

[0124] S20103: Substitute the first relationship in step S20101 into the damage amount formula in step S20102 to obtain the structural damage amount D:

[0125]

[0126] When the response is a narrowband vibration, N0 = E(p), where E(p) is the expectation of the peak frequency of the random response signal. The peak frequency of the narrowband process is close to the zero crossing frequency, and the expectation is expressed as:

[0127]

[0128] In the formula, ω n is the structural natural frequency;

[0129] S202: Using the one-to-one correspondence between the peaks and troughs of the narrowband random process, the stress amplitude p(σ) obeying the Rayleigh distribution is obtained as the probability density function of the stress amplitude:

[0130]

[0131] In the formula, σ RMS is the root mean square value of the structural dynamic response force,

[0132] S203: Substitute the stress amplitude p(σ) calculated in step S202 as the probability density function of the stress amplitude into the expression of the structural damage amount D calculated in step S201 based on the linear cumulative damage Miner criterion, and assume that the constant amplitude SN curve has a linear characteristic on the double logarithmic graph, and obtain the damage accumulation D under the narrow band action NB (T):

[0133]

[0134] S204:D NB When (T)=1, fatigue failure occurs, and the fatigue life, i.e., the narrow-band response structure life prediction model, is:

[0135]

[0136] Specifically, in this exemplary embodiment, a narrowband response structure life prediction model is established based on the narrowband approximation method. Based on the SN curve and the linear cumulative damage Miner criterion, the damage amount of the structure within time T is obtained (step S201); using the one-to-one correspondence between the peaks and troughs of the narrowband random process, its stress amplitude p(σ) is obtained to obey the Rayleigh distribution (step S202). Substitute the probability density function p(σ) of the narrowband stress amplitude into the general expression of damage obtained based on the linear cumulative damage Miner criterion, and assume that the constant amplitude SN curve has linear characteristics on the double logarithmic graph, and obtain the fatigue life of the structure whose vibration response is a narrowband process (steps S203 and S204):

[0137]

[0138] In the formula, ω n is the structural natural frequency, σ RMS is the RMS value of the structural dynamic response force, b and A are constants, which are obtained by fitting the material fatigue performance test data.

[0139] Among them, the vibration signal of the acceleration sensor installed at a specific point of the equipment is collected and processed in the frequency domain to obtain the power spectrum density of the response stress and its root mean square value. At the same time, the SN curve parameters are obtained by performing fatigue tests on the equipment materials, and the above-mentioned vibration fatigue life prediction method based on power spectrum density can be used to evaluate the fatigue failure of the equipment.

[0140] More preferably, in an exemplary embodiment, the establishment of a broadband response life prediction model is specifically based on α 0.75 Methods A broadband response structure life prediction model is established, including:

[0141] S205: Introducing parameter α 0.75 , correct the damage accumulation D under narrowband action calculated in step S203 NB , we can get the damage accumulation D under broadband action WB :

[0142]

[0143] The parameter α 0.75 Defined as:

[0144]

[0145] In the formula, m0, m 0.75 and m 1.5 They are the 0th order spectral moment, 0.75th order spectral moment and 1.5th order spectral moment of the random process respectively;

[0146] S206:D WBWhen (T) = 1, fatigue failure occurs, and the fatigue life is the broadband response structure life prediction model:

[0147]

[0148] Among them, the parameter β is:

[0149]

[0150] Specifically, in this exemplary embodiment, based on α 0.75 The method establishes a broadband response structure life prediction model. The broadband random process is characterized by a wide range of response spectrum and non-centralized response spectrum, which is different from the amplitude and frequency characteristics of the narrowband process. In this case, the use of the narrowband approximation method for calculation and analysis will overestimate the damage degree of the given broadband random process. The coefficient α is introduced 0.75 , the result estimated by the narrowband approximation method (step S203) is adjusted and corrected to compensate for the randomness of the narrowband approximation method for the broadband process. 0.75 Defined as:

[0151]

[0152] In the formula, m0, m 0.75 and m 1.5 They are the 0th, 0.75th and 1.5th order moments of the random process respectively.

[0153] The fatigue life of the structure whose vibration response is a broadband process is obtained as follows:

[0154]

[0155] In order to facilitate the probability fitting of the spectral moment and the statistical modeling of fatigue life based on the MC method (i.e., steps S3 to S5), the parameter β is defined as

[0156]

[0157] More preferably, in an exemplary embodiment, in step S3, for a narrowband random process, the random vibration response spectrum moment is regarded as a group of random samples for statistical analysis, and it is assumed that it obeys a corresponding distribution, a statistical method is used to obtain fitting distribution parameters of the corresponding distribution and obtain a fitting distribution, and a goodness of fit test is performed on the fitting distribution, including:

[0158] S301: For a narrowband random process, assuming σ RMS is a random variable and follows a normal distribution, that is, σ RMS ~N(μ gsg ,σ gsg ), we get σ RMS Distribution parameter μ gsg and σgsg for;

[0159] S302: Use the Kolmogorov-Smirnov method to calculate σ RMS ~N(μ gsg ,σ gsg ) to conduct the test, and accept the null hypothesis when the significance level is 5%, that is, σ RMS Satisfy σ RMS ~N(μ gsg ,σ gsg );

[0160] In step S4, for the narrow-band random process, Monte Carlo sampling is performed on the fitting distribution that passes the goodness of fit test to obtain a sampling result, including:

[0161] S401: Based on the σ obtained in step S302 RMS Fitting distribution σ RMS ~N(μ gsg ,σ gsg ) for σ RMS Perform n samplings and obtain n sampling results.

[0162] More preferably, in an exemplary embodiment, in step S3, for a broadband random process, the random vibration response spectrum moment is regarded as a group of random samples for statistical analysis, and it is assumed that it obeys a corresponding distribution, a statistical method is used to obtain fitting distribution parameters of the corresponding distribution and obtain a fitting distribution, and a goodness of fit test is performed on the fitting distribution, including:

[0163] S303: For a broadband random process, according to the parameter β of the random process, similarly, assuming that β obeys a normal distribution β~N(μ β ,σ β ), and obtain the β distribution parameter μ β and σ β ;

[0164] S304: Use the Kolmogorov-Smirnov method to calculate β~N(μ β ,σ β ) to conduct the test, and accept the null hypothesis when the significance level is 5%, that is, β~N(μ β ,σ β );

[0165] In step S4, for the broadband random process, Monte Carlo sampling is performed on the fitting distribution that passes the goodness of fit test to obtain a sampling result, including:

[0166] S401: Fitting distribution β~N(μ β ,σ β) Sample β n times and obtain n sampling results.

[0167] Specifically, in this exemplary embodiment, the random vibration response spectrum moments of a group of products are regarded as a group of random samples for statistical analysis, which are assumed to obey a certain distribution, and then the distribution parameters are obtained by using statistical methods, and a goodness of fit test is performed.

[0168] For a narrowband random process, in step S301, the root mean square value of the response acceleration σ RMS Conduct hypothesis testing analysis and obtain σ RMS The distribution and distribution parameters of the material and load are taken into account. Considering the uncertainty of materials and loads and the differences in samples, it is reasonable to assume that σ RMS is a random variable and follows a normal distribution, that is, σ RMS ~N(μ gsg ,σ gsg ). For a broadband random process, in step S303, the random process parameter β is calculated, and a hypothesis test analysis is performed on the parameter β to obtain the distribution and distribution parameters that β obeys. Similarly, it can be assumed that β~N(μ β ,σ β ).

[0169] In step S302 and step S304, the Kolmogorov-Smirnov (KS) goodness-of-fit test is used to determine whether the original distribution hypothesis is rejected at a certain significance level.

[0170] For narrowband and broadband random process responses, the Monte Carlo method is used to generate parameters σ RMS and β are a set of random numbers. The root mean square value of the narrowband random process σ RMS The probability density function is f gsg (σ RMS ), the probability density function of the broadband random process parameter β is f β (β), according to the established narrowband lifetime prediction model based on the narrowband approximation method and the 0.75 The broadband life prediction model based on the method, fatigue life distribution characteristics f NB (T NB ), f WB (T WB ) and the response distribution characteristics f gsg (σ RMS ), f β (β) have a certain relationship, which is similar to a function mapping relationship, which can be calculated based on the root mean square value of the response acceleration σ RMS Or parameter β is used to predict the equipment fatigue life distribution.

[0171] Therefore, for the narrowband random process, in step S401, according to σ RMSFitting distribution σ RMS ~N(μ gsg ,σ gsg ) for σ RMS Perform n = 1000 samplings, and the sampling results are as follows Figure 2 As shown; in step S402, for a broadband random process, according to the β fitting distribution β~N(μ β ,σ β ) Sampling β for n = 1000 times, the sampling results are as follows Figure 3 shown.

[0172] More preferably, in an exemplary embodiment, in step S5, the sampling result is input into the life prediction model to obtain a fatigue life sample, and after the fatigue life sample passes the goodness test, a fatigue life empirical distribution function is obtained, including:

[0173] S501: inputting the n sampling results obtained in step S4 into the life prediction model obtained in step S2 to obtain n fatigue life samples;

[0174] S502: using the Kolmogorov-Smirnov goodness of fit test, the n fatigue life samples obtained in step S501 are sequentially tested for log-normal distribution, Weibull distribution and exponential distribution hypothesis, and when a 5% significance level is obtained, it is considered that the life of the narrow-band / broadband vibration response product obeys the log-normal distribution;

[0175] S503: Plotting fatigue life empirical distribution functions corresponding to the n sampling results obtained in step S4.

[0176] Specifically, in this exemplary embodiment, a life sample is obtained according to the established life prediction model, and probability fitting and distribution hypothesis testing are performed based on the sample data.

[0177] For narrowband random processes, σ is calculated according to the narrowband lifetime prediction model based on the narrowband approximation method. RMS The fatigue life corresponding to the 1000 sampling results (step S501) is calculated as follows: Figure 4 As shown; 1000 fatigue life samples are tested for normal distribution, Weibull distribution and exponential distribution hypothesis in turn; the Kolmogorov-Smirnov (KS) goodness of fit test is used to determine whether the original distribution hypothesis is rejected at a certain significance level (step S502). When the significance level is 5%, it can be considered that the narrow-band vibration response product life obeys the log-normal distribution, that is, in:

[0178]

[0179] Plot the empirical distribution function of narrow-band fatigue life corresponding to 1000 sampling results and the hypothesis test prediction results, such as Figure 5 shown.

[0180] For broadband random processes, according to the 0.75 The broadband life prediction model of the method calculates the fatigue life corresponding to the sampling results of β1000 times (step S501) and the calculation results are as follows: Figure 6 As shown; Similarly, the Kolmogorov-Smirnov (KS) goodness of fit test was used to test the log-normal distribution, Weibull distribution and exponential distribution hypothesis of 1000 fatigue life samples in turn. When the 5% significance level was obtained, it can be considered that the life of the broadband vibration response product obeys the log-normal distribution, that is, in

[0181]

[0182] Plot the empirical distribution function of broadband fatigue life corresponding to 1000 sampling results and the hypothesis test prediction results, such as Figure 7 shown.

[0183] With the same inventive concept as the above exemplary embodiment, another exemplary embodiment of the present invention provides an electronic device, comprising a storage unit and a processing unit, wherein the storage unit stores computer instructions that can be executed on the processing unit, and when the processing unit executes the computer instructions, the processing unit executes the following Figure 1 The steps of the method.

[0184] The electronic device is presented in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: the at least one processing unit, the at least one storage unit, and a bus connecting different system components (including the storage unit and the processing unit).

[0185] The storage unit stores program codes, which can be executed by the processing unit, so that the processing unit performs the steps according to various exemplary embodiments of the present invention described in the above “Exemplary Method” section of this specification. For example, the processing unit can perform the following steps: Figure 1 The method shown in .

[0186] The storage unit may include a readable medium in the form of a volatile memory unit, such as a random access memory unit (RAM) and / or a cache memory unit, and may further include a read-only memory unit (ROM).

[0187] The storage unit may also include a program / utility having a set (at least one) of program modules, such program modules including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0188] The bus may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0189] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may communicate with one or more devices that enable a user to interact with the electronic device, and / or may communicate with any device that enables the electronic device to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface. Furthermore, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter. The network adapter communicates with other modules of the electronic device through a bus. It should be understood that other hardware and / or software modules may be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0190] Through the above description, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to this exemplary embodiment can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to this exemplary embodiment.

[0191] With the same inventive concept as the above exemplary embodiment, another exemplary embodiment of the present invention provides a storage medium having computer instructions stored thereon, wherein the computer instructions are executed as follows when they are executed: Figure 1 The steps of the method.

[0192] Based on such understanding, the technical solution of this embodiment can essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product (program product). The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention.

[0193] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0194] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0195] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0196] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0197] Obviously, the above embodiments are merely examples for clear explanation, and are not intended to limit the implementation methods. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from them are still within the protection scope of the invention.

Claims

1. A vibration fatigue life prediction method considering the influence of random factors, characterized by: include: S1: Calculate the power spectrum of the random vibration response time domain signal to obtain the random vibration response spectrum moment, and further use the random vibration response spectrum moment to perform bandwidth analysis to determine whether the vibration response type belongs to narrowband response or broadband response; S2: Establish a life prediction model corresponding to the vibration response type; S3: treating the random vibration response spectrum moment as a group of random samples for statistical analysis, assuming that it obeys the corresponding distribution, using statistical methods to obtain the fitting distribution parameters of the corresponding distribution and obtain the fitting distribution, and performing a goodness of fit test on the fitting distribution; S4: Perform Monte Carlo sampling on the fitting distribution that passes the goodness of fit test to obtain the sampling results; S5: Input the sampling results into the life prediction model to obtain fatigue life samples, and obtain fatigue life empirical distribution function after the fatigue life samples pass the goodness of fit test.

2. The vibration fatigue life prediction method considering the influence of random factors according to claim 1 is characterized in that: In step S1, the power spectrum of the random vibration response time domain signal is calculated to obtain the random vibration response spectral moment, including: S101: Based on the density of the power spectrum of the random vibration response time domain signal, the i-th order spectral moment m of the random process is obtained. i for: Where G(f) is the power spectrum density; f is the frequency; In step S1, the use of random vibration response spectrum moment to perform bandwidth analysis to determine whether the vibration response type belongs to narrowband response or broadband response includes: S102: Calculate the 0th order spectral moment m0, the 2nd order spectral moment m2, and the 4th order spectral moment m4 respectively; S103: Calculate the irregularity factor γ of the random process using the 0th order spectral moment m0, the 2nd order spectral moment m2, and the 4th order spectral moment m4: S104: Calculate the bandwidth coefficient ε describing the bandwidth of the random process using the irregular factor γ: When the bandwidth coefficient ε approaches 0, the random process conforms to a narrowband random process, and the vibration response type is judged to be a narrowband response; when the bandwidth coefficient ε approaches 1, the random process corresponds to a broadband random process, and the vibration response type is judged to be a broadband response.

3. The vibration fatigue life prediction method considering the influence of random factors according to claim 2 is characterized in that: In step S2, the establishing of the life prediction model corresponding to the vibration response type includes establishing a narrowband response life prediction model and establishing a broadband response life prediction model.

4. The vibration fatigue life prediction method considering the influence of random factors according to claim 3 is characterized in that: The method of establishing a narrowband response life prediction model is to establish a narrowband response structure life prediction model based on a narrowband approximation method, specifically including: S201: Based on the SN curve and the linear cumulative damage Miner criterion, the damage amount of the structure within time T is obtained, including: S20101: Based on the SN curve, fatigue life is predicted, and the stress amplitude σ and fatigue life N σ The following first relationship exists: s b N σ =A In the formula, b and A are constants, which are obtained by fitting the material fatigue performance test data; S20102: Based on the linear cumulative damage Miner criterion, the structural damage D of the structure within the action time T is: In the formula, The amplitude is σ i The number of times the stress acts at time T, is the stress amplitude σ i Fatigue life under, N is stress σ i The average number of cycles within the range of Δσ, p(σ i ) represents the stress amplitude σ i The probability density function of The average number of cycles in the stress range is obtained by the following formula: N=N0T Where N0 is the stress per unit time σ i The average number of cycles within the range Δσ; S20103: Substitute the first relationship in step S20101 into the damage amount formula in step S20102 to obtain the structural damage amount D: When the response is a narrowband vibration, N0 = E(p), where E(p) is the expectation of the peak frequency of the random response signal. The peak frequency of the narrowband process is close to the zero crossing frequency, and the expectation is expressed as: In the formula, ω n is the structural natural frequency; S202: Using the one-to-one correspondence between the peaks and troughs of the narrowband random process, the stress amplitude p(σ) obeying the Rayleigh distribution is obtained as the probability density function of the stress amplitude: In the formula, σ RMS is the root mean square value of the structural dynamic response force, S203: Substitute the stress amplitude p(σ) calculated in step S202 as the probability density function of the stress amplitude into the expression of the structural damage amount D calculated in step S201 based on the linear cumulative damage Miner criterion, and assume that the constant amplitude SN curve has a linear characteristic on the double logarithmic graph, and obtain the damage accumulation D under the narrow band action NB (T): S204:D NB When (T)=1, fatigue failure occurs, and the fatigue life, i.e., the narrow-band response structure life prediction model, is:

5. The vibration fatigue life prediction method considering the influence of random factors according to claim 4 is characterized in that: The broadband response life prediction model is specifically established based on α 0.75 Methods A broadband response structure life prediction model is established, including: S205: Introducing parameter α 0.75 , correct the damage accumulation D under narrowband action calculated in step S203 NB , we can get the damage accumulation D under broadband action WB : The parameter α 0.75 Defined as: In the formula, m0, m 0.75 and m 1.5 They are the 0th order spectral moment, 0.75th order spectral moment and 1.5th order spectral moment of the random process respectively; S206:D WB When (T) = 1, fatigue failure occurs, and the fatigue life is the broadband response structure life prediction model: Among them, the parameter β is:

6. The vibration fatigue life prediction method considering the influence of random factors according to claim 4 is characterized in that: In step S3, for a narrowband random process, the random vibration response spectrum moment is regarded as a group of random samples for statistical analysis, and it is assumed that it obeys the corresponding distribution. A statistical method is used to obtain the fitting distribution parameters of the corresponding distribution and obtain the fitting distribution, and a goodness of fit test is performed on the fitting distribution, including: S301: For a narrowband random process, assuming σ RMS is a random variable and follows a normal distribution, that is, σ RMS ~N(μ gsg ,σ gsg ), we get σ RMS Distribution parameter μ gsg and σ gsg for; S302: Use the Kolmogorov-Smirnov method to calculate σ RMS ~N(μ gsg ,σ gsg ) to conduct the test, and accept the null hypothesis when the significance level is 5%, that is, σ RMS Satisfy σ RMS ~N(μ gsg ,σ gsg ); In step S4, for the narrow-band random process, Monte Carlo sampling is performed on the fitting distribution that passes the goodness of fit test to obtain a sampling result, including: S401: Based on the σ obtained in step S302 RMS Fitting distribution σ RMS ~N(μ gsg ,σ gsg ) for σ RMS Perform n samplings and obtain n sampling results.

7. The vibration fatigue life prediction method considering the influence of random factors according to claim 5 is characterized in that: In step S3, for a broadband random process, the random vibration response spectrum moment is regarded as a group of random samples for statistical analysis, and it is assumed that it obeys the corresponding distribution. A statistical method is used to obtain the fitting distribution parameters of the corresponding distribution and obtain the fitting distribution, and a goodness of fit test is performed on the fitting distribution, including: S303: For a broadband random process, according to the parameter β of the random process, similarly, assuming that β obeys a normal distribution β~N(μ β ,σ β ), and obtain the β distribution parameter μ β and σ β ; S304: Use the Kolmogorov-Smirnov method to calculate β~N(μ β ,σ β ) to conduct the test, and accept the null hypothesis when the significance level is 5%, that is, β~N(μ β ,σ β ); In step S4, for the broadband random process, Monte Carlo sampling is performed on the fitting distribution that passes the goodness of fit test to obtain a sampling result, including: S401: Fitting distribution β~N(μ β ,σ β ) Sample β n times and obtain n sampling results.

8. A vibration fatigue life prediction method considering the influence of random factors according to claim 6 or 7, characterized in that: In step S5, the sampling result is input into the life prediction model to obtain a fatigue life sample, and after the fatigue life sample passes the goodness test, a fatigue life empirical distribution function is obtained, including: S501: inputting the n sampling results obtained in step S4 into the life prediction model obtained in step S2 to obtain n fatigue life samples; S502: using the Kolmogorov-Smirnov goodness of fit test, the n fatigue life samples obtained in step S501 are sequentially tested for log-normal distribution, Weibull distribution and exponential distribution hypothesis, and when a 5% significance level is obtained, it is considered that the life of the narrow-band / broadband vibration response product obeys the log-normal distribution; S503: Plotting fatigue life empirical distribution functions corresponding to the n sampling results obtained in step S4.

9. An electronic device, comprising a storage unit and a processing unit, wherein the storage unit stores computer instructions that can be executed on the processing unit, wherein: When the processing unit runs the computer instructions, the steps of the method according to any one of claims 1 to 8 are performed.

10. A storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed, the steps of the method according to any one of claims 1 to 8 are executed.

Images