Time domain direct fatigue analysis method, device, terminal and medium based on EMD

Abstract

The present application provides a time-domain direct fatigue analysis method, device, terminal and medium based on EMD, the method comprising: collecting a load signal x(t) received by a mechanical structure; performing empirical mode decomposition (EMD) on the collected load signal x(t) to obtain each characteristic mode function (IMF) component; performing preliminary screening on each IMF component to remove false components; further screening each IMF component preliminarily screened out to screen out IMF components whose contribution to the total energy value of the original load signal is greater than a predetermined value; accumulating the fatigue damage caused by each IMF component to the mechanical structure to obtain the total fatigue damage of the mechanical structure under the action of the original load signal. The present invention can avoid energy leakage and waveform distortion of the load signal after fast Fourier transform, well retain the amplitude and frequency of the useful signal, and improve the ability to identify subtle features of the original load signal.

Description

Technical Field

[0001] The present application relates to the technical field of fatigue simulation analysis of structural parts, and in particular to a time-domain direct fatigue analysis method, device, terminal and medium based on EMD. Background Art

[0002] Fatigue failure of mechanical structures has become an important factor in many engineering fields related to the safety and cost-effectiveness of machinery or structures, and sometimes even to people's lives. The American Society for Testing and Materials defines fatigue as the development process of local, permanent structural changes in materials that are subjected to disturbance stress at one or more points and form cracks or complete fractures after sufficient cyclic disturbances.

[0003] In the process of fatigue analysis of mechanical structures, it is often necessary to perform Fourier transform on the load signal. However, the spectrum obtained by Fourier transform of the load signal will inevitably leak, and the waveform will be distorted to varying degrees, which will affect the accuracy of fatigue analysis and cause large errors in the calculation results. Summary of the invention

[0004] In view of this, the present application provides a time-domain direct fatigue analysis method, device, terminal and medium based on EMD, aiming to achieve a structural fatigue analysis solution with smaller error and higher accuracy.

[0005] To achieve the above objectives, the technical solutions adopted in this application are as follows:

[0006] In a first aspect, the present application provides a time domain direct fatigue analysis method based on EMD, comprising:

[0007] Step S1: collecting the load signal x(t) exerted on the mechanical structure;

[0008] Step S2: performing empirical mode decomposition (EMD) on the collected load signal x(t) to obtain each characteristic mode function (IMF) component of the load signal;

[0009] Step S3: construct a random white noise sample, establish a logarithmic confidence interval of the energy density of the white noise IMF based on the statistical characteristics of the white noise, perform a preliminary screening of the IMF components of the load signal, and remove false components;

[0010] Step S4: further screening the IMF components initially screened out to screen out the IMF components whose contribution to the total energy value of the original load signal is greater than a predetermined value;

[0011] Step S5: using the IMF components of the load signal screened out in step S4 as input loads for fatigue analysis, and calculating fatigue damage caused by each order of IMF components to the mechanical structure;

[0012] Step S6: Accumulate the fatigue damage caused by each IMF component obtained in step S5 to the mechanical structure to obtain the total fatigue damage of the mechanical structure under the action of the original load signal.

[0013] Optionally, step S2 further includes:

[0014] 2-1: Determine all the extreme points of the collected original load signal x(t), and use cubic spline function interpolation to form the upper envelope curve u of the data for all the maximum and minimum points 1 (t) and the lower envelope curve u 2 (t), and take u 1 (t) and u 2 The average value of (t) 1 (t), that is

[0015] m 1 (t)=(u 1 (t)+u 2 (t)) / 2 (1)

[0016] 2-2: Let h 1 (t) = x(t) - m 1 (t), if h 1 (t) satisfies two conditions of IMF at the same time: ① The number of zero crossings and the number of extreme points are equal or differ by at most 1 in the entire time history; ② The upper envelope defined by the local maximum and the lower envelope defined by the local minimum are locally symmetrical about the time axis, that is, the mean of the upper envelope defined by the local maximum and the lower envelope defined by the local minimum is 0, then h 1 (t) is the first-order IMF. If it is not satisfied, h 1 (t) is regarded as the new x(t), m 11 (t) is the mean of its upper and lower envelopes, and h 11 (t) = h 1 (t)-m 11 (t), if h 11 (t) is still not satisfied, repeat the process k times, and get

[0017] h 1k (t) = h 1(k-1) (t)-m 1k (t) (2)

[0018] If h 1k (t) and h 1(k-1) If the standard deviation SD of (t) is within the predetermined range, the process is stopped. 1k (t) is the first-order IMF of the original load signal x(t), denoted by c 1 (t) = h 1k(t), where the standard deviation SD is calculated as:

[0019]

[0020] Where T is the total time length of the load signal x(t);

[0021] 2-3: Let r 1 (t) = x(t) - c 1 (t), r 1 (t) is regarded as the new x(t), and step 2-2 is repeated to obtain the second-order IMF, denoted as c 2 (t), and then the other order IMFs are obtained by analogy, which are denoted as c 3 (t)……c i (t), until r i (t) is a monotonic function and the IMF can no longer be separated.

[0022] Optionally, step S3 further includes:

[0023] 3-1: Perform Fourier transform on each order IMF component obtained in step S2 to obtain the frequency spectrum of each order IMF component;

[0024] 3-2: Calculate the energy density logarithm lnE of each order IMF component i And the average period logarithm value , the calculation formulas are:

[0025] oeLh i =ln|C i (f)| 2 (4)

[0026]

[0027] In the formula, C i (f) is the IMF component c i The Fourier transform spectrum of (t), S lnT',i c i (t) Fourier transform spectrum of lnT', T' is the IMF component c i (t) period of Fourier transform;

[0028] 3-3: Construct several normalized random white noise samples with the same length as the original load signal x(t);

[0029] 3-4: Perform EMD decomposition on the white noise to obtain IMF components of each order of white noise, and establish the energy density logarithm lnE of each order of white noise IMF components at a specific confidence level according to the statistical characteristics of white noise j The distribution interval template, lnEj The confidence interval calculation formula is:

[0030]

[0031] In the formula, is the average period logarithm of the Fourier spectrum of each order white noise IMF component, Z α / 2 You can check the standard normal distribution table, N is the number of white noise samples, that is, the number of time series, and j represents different white noise IMF components;

[0032] 3-5: The energy density logarithm of each IMF component decomposed from the original load signal x(t) is lnE i and the average period logarithm Compare with the confidence interval established in step 3-4 to filter out the true components and eliminate useless false components.

[0033] Optionally, step S4 further includes:

[0034] 4-1: Calculate the energy value of each order IMF component screened in step S3 and the energy value E of the original load signal x , the calculation formulas are:

[0035]

[0036]

[0037] 4-2: Calculate the contribution β of each order IMF component screened in step S3 to the total energy value of the original load signal x(t) i , the calculation formula is

[0038]

[0039] 4-3: According to the contribution calculated in step 4-2, the IMF components whose contribution to the total energy value of the original load signal x(t) is greater than a predetermined value are screened out.

[0040] In a second aspect, the present application provides a time domain direct fatigue analysis device based on EMD, comprising:

[0041] An acquisition module, used for acquiring a load signal x(t) received by the mechanical structure;

[0042] A decomposition module, used for performing empirical mode decomposition (EMD) on the collected load signal x(t) to obtain each characteristic mode function (IMF) component of the load signal;

[0043] The first screening module is used to construct a random white noise sample, establish a logarithmic confidence interval of the energy density of the white noise IMF based on the statistical characteristics of the white noise, perform a preliminary screening of the IMF components of the load signal, and remove false components;

[0044] A second screening module is used to further screen the IMF components initially screened out to screen out the IMF components whose contribution to the total energy value of the original load signal is greater than a predetermined value;

[0045] A first calculation module, used for taking the IMF components of the load signal screened by the second screening module as input loads for fatigue analysis, and respectively calculating fatigue damage caused by each order of IMF components to the mechanical structure;

[0046] The second calculation module is used to accumulate the fatigue damage caused to the mechanical structure by each IMF component obtained by the first calculation module to obtain the total fatigue damage of the mechanical structure under the action of the original load signal.

[0047] Optionally, the decomposition module is specifically used to perform the following steps:

[0048] a1: Determine all the extreme points of the collected original load signal x(t), and use cubic spline function interpolation to form the upper envelope curve u of the data for all the maximum and minimum points 1 (t) and the lower envelope curve u 2 (t), and take u 1 (t) and u 2 The average value of (t) 1 (t), that is

[0049] m 1 (t)=(u 1 (t)+u 2 (t)) / 2 (1)

[0050] a2: Let h 1 (t) = x(t) - m 1 (t), if h 1 (t) satisfies two conditions of IMF at the same time: ① The number of zero crossings and the number of extreme points are equal or differ by at most 1 in the entire time history; ② The upper envelope defined by the local maximum and the lower envelope defined by the local minimum are locally symmetrical about the time axis, that is, the mean of the upper envelope defined by the local maximum and the lower envelope defined by the local minimum is 0, then h 1 (t) is the first-order IMF. If it is not satisfied, h 1 (t) is regarded as the new x(t), m 11 (t) is the mean of its upper and lower envelopes, and h 11 (t) = h 1 (t)-m 11 (t), if h11 (t) is still not satisfied, repeat the process k times, and get

[0051] h 1k (t) = h 1(k-1) (t)-m 1k (t) (2)

[0052] If h 1k (t) and h 1(k-1) If the standard deviation SD of (t) is within the predetermined range, the process is stopped. 1k (t) is the first-order IMF of the original load signal x(t), denoted by c 1 (t) = h 1k (t), where the standard deviation SD is calculated as:

[0053]

[0054] Where T is the total time length of the load signal x(t);

[0055] a3: Let r 1 (t) = x(t) - c 1 (t), r 1 (t) is regarded as the new x(t), and step a2 is repeated to obtain the second-order IMF, denoted as c 2 (t), and then the other order IMFs are obtained by analogy, which are denoted as c 3 (t)……c i (t), until r i (t) is a monotonic function and the IMF can no longer be separated.

[0056] Optionally, the first screening module is specifically configured to perform the following steps:

[0057] b1: Perform Fourier transform on each order IMF component obtained by the decomposition module to obtain the spectrum of each order IMF component;

[0058] b2: Calculate the energy density logarithm lnE of each order IMF component i And the average period logarithm -

[0059] Value lnT i , the calculation formulas are:

[0060] oeLh i =ln|C i (f)| 2 (4)

[0061]

[0062] In the formula, Ci (f) is the IMF component c i The Fourier transform spectrum of (t), S lnT',i c i (t) Fourier transform spectrum of lnT', T' is the IMF component c i (t) period of Fourier transform;

[0063] b3: Construct several normalized random white noise samples with the same length as the original load signal x(t);

[0064] b4: Perform EMD decomposition on the white noise to obtain the IMF components of each order of white noise. According to the statistical characteristics of white noise, establish the energy density logarithm lnE of each order of white noise IMF components at a specific confidence level j The distribution interval template, lnE j The confidence interval calculation formula is:

[0065]

[0066] In the formula, is the average period logarithm of the Fourier spectrum of each order white noise IMF component, Z α / 2 You can check the standard normal distribution table, N is the number of white noise samples, that is, the number of time series, and j represents different white noise IMF components;

[0067] b5: The energy density logarithm of each IMF component decomposed from the original load signal x(t) lnE i and the average period logarithm Compare with the confidence interval established in step 3-4 to filter out the true components and eliminate useless false components.

[0068] Optionally, the second screening module is specifically configured to perform the following steps:

[0069] c1: respectively obtain the energy value of each order IMF component screened by the first screening module and the energy value E of the original load signal x , the calculation formulas are:

[0070]

[0071]

[0072] c2: Calculate the contribution β of each order IMF component screened by the first screening module to the total energy value of the original load signal x(t) i , the calculation formula is

[0073]

[0074] c3: According to the contribution calculated in step c2, select the IMF components whose contribution to the total energy value of the original load signal x(t) is greater than a predetermined value.

[0075] In a third aspect, an embodiment of the present application further provides a terminal, including: a processor, a memory, and a communication unit;

[0076] The memory stores machine-readable instructions executable by the processor, and when the device is running, the processor communicates with the memory through the communication unit;

[0077] The processor executes the machine-readable instructions to perform the methods described in the above aspects.

[0078] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the methods described in the above aspects are executed.

[0079] The beneficial effects of this application are:

[0080] 1. This application applies the EMD method to fatigue analysis, expanding the application field of the EMD method. The method of this application can be used for fatigue analysis of various mechanical structures;

[0081] 2. This application uses EMD decomposition on the original load signal, and screens the IMF based on the statistical characteristics of white noise and energy contribution, which can avoid energy leakage and waveform distortion of the load signal after fast Fourier transform, well preserve the amplitude and frequency of the useful signal, and improve the ability to recognize subtle features of the original load signal;

[0082] 3. This application uses the EMD method to decompose the measured signal, which can remove the interference signal in the original load signal and characterize the main features of the measured signal, thus avoiding cumbersome signal processing, speeding up the calculation of fatigue damage of mechanical structures and improving the efficiency of fatigue analysis;

[0083] 4. This application can improve the signal-to-noise ratio of the signal and better suppress the influence of interference signals on fatigue analysis results. The fatigue damage results calculated using this method are more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0085] Figure 1 This is a flow chart of the time domain direct fatigue analysis method based on EMD for this application;

[0086] Figure 2 This is a schematic diagram of the time domain direct fatigue analysis method based on EMD in this application;

[0087] Figure 3 Flowchart for performing EMD for this application;

[0088] Figure 4 This is a schematic diagram of screening the IMF components of the original load signal based on the statistical characteristics of white noise in this application;

[0089] Figure 5 This is a structural block diagram of the time domain direct fatigue analysis device based on EMD in this application. DETAILED DESCRIPTION

[0090] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0091] like Figure 1 and 2 As shown, the first aspect of the present application provides a time domain direct fatigue analysis method based on EMD, comprising:

[0092] Step S1: collecting the load signal x(t) exerted on the mechanical structure;

[0093] The load signal x(t) can be collected through actual measurement or by using a multi-body dynamics model and virtual iteration.

[0094] Step S2: performing empirical mode decomposition (EMD) on the collected load signal x(t) to obtain each characteristic mode function (IMF) component of the load signal;

[0095] Empirical Mode Decomposition (EMD) is a new adaptive signal time-frequency processing method, which is particularly suitable for the analysis and processing of nonlinear and non-stationary signals. The key to this method is empirical mode decomposition, which can decompose complex signals into a finite number of intrinsic mode functions (IMFs). The decomposed IMF components contain local characteristic signals of the original signal at different time scales.

[0096] As an optional implementation, Figure 3 As shown, the step S2 further comprises:

[0097] 2-1: Determine all the extreme points of the collected original load signal x(t), and use cubic spline function interpolation to form the upper envelope curve u of the data for all the maximum and minimum points 1 (t) and the lower envelope curve u 2 (t), and take u 1 (t) and u 2 The average value of (t) 1 (t), that is

[0098] m 1 (t)=(u 1 (t)+u 2 (t)) / 2 (1)

[0099] 2-2: Let h 1 (t) = x(t) - m 1 (t), if h 1 (t) satisfies two conditions of IMF at the same time: ① The number of zero crossings and the number of extreme points are equal or differ by at most 1 in the entire time history; ② The upper envelope defined by the local maximum and the lower envelope defined by the local minimum are locally symmetrical about the time axis, that is, the mean of the upper envelope defined by the local maximum and the lower envelope defined by the local minimum is 0, then h 1 (t) is the first-order IMF. If it is not satisfied, h 1 (t) is regarded as the new x(t), m 11 (t) is the mean of its upper and lower envelopes, and h 11 (t) = h 1 (t)-m 11 (t), if h 11 (t) is still not satisfied, repeat the process k times, and get

[0100] h 1k (t) = h 1(k-1) (t)-m 1k (t) (2)

[0101] If h 1k (t) and h1(k-1) If the standard deviation SD of (t) is within the predetermined range, the process is stopped. 1k (t) is the first-order IMF of the original load signal x(t), denoted by c 1 (t) = h 1k (t), where the standard deviation SD is calculated as:

[0102]

[0103] Where T is the total time length of the load signal x(t);

[0104] Specifically, the predetermined range of the standard deviation SD is generally 0.2≥SD≥0.3, but is not limited thereto, and the predetermined range may also be adjusted differently according to actual conditions.

[0105] 2-3: Let r 1 (t) = x(t) - c 1 (t), r 1 (t) is regarded as the new x(t), and step 2-2 is repeated to obtain the second-order IMF, denoted as c 2 (t), and then the other order IMFs are obtained by analogy, which are denoted as c 3 (t)……c i (t), until r i (t) is a monotonic function and the IMF can no longer be separated.

[0106] This application uses the EMD method to decompose the measured signal, which can avoid tedious signal processing, speed up the calculation of fatigue damage of mechanical structures, and improve the efficiency of fatigue analysis.

[0107] Step S3: construct a random white noise sample, establish a logarithmic confidence interval of the energy density of the white noise IMF based on the statistical characteristics of the white noise, perform a preliminary screening of the IMF components of the load signal, and remove false components;

[0108] As an optional implementation, step S3 further includes:

[0109] 3-1: Perform Fourier transform on each order IMF component obtained in step S2 to obtain the frequency spectrum of each order IMF component;

[0110] 3-2: Calculate the energy density logarithm lnE of each order IMF component i And the average period logarithm value The calculation formulas are:

[0111] oeLh i =ln|C i (f)| 2(4)

[0112]

[0113] In the formula, C i (f) is the IMF component c i The Fourier transform spectrum of (t), S lnT',i c i (t) Fourier transform spectrum of lnT', T' is the IMF component c i (t) period of Fourier transform;

[0114] 3-3: Construct several normalized random white noise samples with the same length as the original load signal x(t);

[0115] 3-4: Perform EMD decomposition on the white noise to obtain IMF components of each order of white noise, and establish the energy density logarithm lnE of each order of white noise IMF components at a specific confidence level according to the statistical characteristics of white noise j The distribution interval template, lnE j The confidence interval calculation formula is:

[0116]

[0117] In the formula, is the average period logarithm of the Fourier spectrum of each order white noise IMF component, Z α / 2 The standard normal distribution table can be consulted, where N is the number of white noise samples, that is, the number of time series, and j represents different white noise IMF components; in a specific embodiment, the specific confidence level can be a confidence level of 99%, or other specific values ​​set according to the situation.

[0118] 3-5: The energy density logarithm of each IMF component decomposed from the original load signal x(t) is lnE i and the average period logarithm Compare with the confidence interval established in step 3-4 to filter out the true components and eliminate useless false components.

[0119] The screening criteria used in this step are that the IMF components outside the confidence interval are valid, and those within the confidence interval are false components. The schematic diagram is shown in Figure 4 As shown, in Figure 4 In the above equation, IMF1, IMF2, IMF4 and IMF5 are valid IMF components, while IMF3, IMF6 and IMF7 are useless false components.

[0120] The present application can improve the signal-to-noise ratio of the signal by constructing a confidence interval to screen the true component of the load signal, and better suppress the influence of the interference signal on the fatigue analysis results. The fatigue damage results calculated using this method are more accurate.

[0121] Step S4: further screening the IMF components initially screened out to screen out the IMF components whose contribution to the total energy value of the original load signal is greater than a predetermined value;

[0122] As an optional implementation, step S4 further includes:

[0123] 4-1: Calculate the energy value of each order IMF component screened in step S3 and the energy value E of the original load signal x , the calculation formulas are:

[0124]

[0125]

[0126] 4-2: Calculate the contribution β of each order IMF component screened in step S3 to the total energy value of the original load signal x(t) i , the calculation formula is

[0127]

[0128] 4-3: According to the contribution calculated in step 4-2, the IMF components whose contribution to the total energy value of the original load signal x(t) is greater than a predetermined value are screened out.

[0129] In a specific embodiment, the predetermined value may be 0.1, or other appropriate values, which may be determined according to actual conditions. The IMF component screened out in this step is used as the input load for subsequent fatigue damage calculation.

[0130] Step S5: using the IMF components of the load signal screened out in step S4 as input loads for fatigue analysis, and calculating fatigue damage caused by each order of IMF components to the mechanical structure;

[0131] In this step, the damage caused by each order IMF component to the mechanical structure is calculated separately, denoted as D 1 , D 2 , D 3 …D i …D nThe calculation of the damage caused by each order IMF component to the mechanical structure can be directly performed in finite element analysis software, such as ncode software, etc. The damage calculation method of the mechanical structure can adopt quasi-static method, psd method (self-spectrum method), cpsd method (cross-spectrum method) and modal superposition method, etc. The specific method to be adopted depends on the situation.

[0132] Step S6: Accumulate the fatigue damage caused by each IMF component obtained in step S5 to the mechanical structure to obtain the total fatigue damage of the mechanical structure under the action of the original load signal.

[0133] The damage caused by each order IMF component calculated in step S5 to the mechanical structure is linearly accumulated to obtain the total fatigue damage D of the mechanical structure under the original excitation or original load signal x(t), that is,

[0134]

[0135] The present application adopts EMD decomposition on the original load signal, and screens the IMF based on the statistical characteristics of white noise and the energy contribution, which can avoid the energy leakage and waveform distortion of the load signal after the fast Fourier transform, well preserve the amplitude and frequency of the useful signal, and improve the recognition ability of the subtle features of the original load signal.

[0136] Second, as Figure 5 As shown, the present application provides a time domain direct fatigue analysis device based on EMD, comprising:

[0137] The acquisition module 510 is used to acquire the load signal x(t) received by the mechanical structure;

[0138] A decomposition module 520 is used to perform empirical mode decomposition (EMD) on the collected load signal x(t) to obtain each characteristic mode function (IMF) component of the load signal;

[0139] As an optional implementation, the decomposition module 520 is specifically configured to perform the following steps:

[0140] a1: Determine all the extreme points of the collected original load signal x(t), and use cubic spline function interpolation to form the upper envelope curve u of the data for all the maximum and minimum points 1 (t) and the lower envelope curve u 2 (t), and take u 1 (t) and u 2 The average value of (t) 1 (t), that is

[0141] m 1 (t)=(u 1 (t)+u 2(t)) / 2 (1)

[0142] a2: Let h 1 (t) = x(t) - m 1 (t), if h 1 (t) satisfies two conditions of IMF at the same time: ① The number of zero crossings and the number of extreme points are equal or differ by at most 1 in the entire time history; ② The upper envelope defined by the local maximum and the lower envelope defined by the local minimum are locally symmetrical about the time axis, that is, the mean of the upper envelope defined by the local maximum and the lower envelope defined by the local minimum is 0, then h 1 (t) is the first-order IMF. If it is not satisfied, h 1 (t) is regarded as the new x(t), m 11 (t) is the mean of its upper and lower envelopes, and h 11 (t) = h 1 (t)-m 11 (t), if h 11 (t) is still not satisfied, repeat the process k times, and get

[0143] h 1k (t) = h 1(k-1) (t)-m 1k (t) (2)

[0144] If h 1k (t) and h 1(k-1) If the standard deviation SD of (t) is within the predetermined range, the process is stopped. 1k (t) is the first-order IMF of the original load signal x(t), denoted by c 1 (t) = h 1k (t), where the standard deviation SD is calculated as:

[0145]

[0146] Where T is the total time length of the load signal x(t);

[0147] a3: Let r 1 (t) = x(t) - c 1 (t), r 1 (t) is regarded as the new x(t), and step a2 is repeated to obtain the second-order IMF, denoted as c 2 (t), and then the other order IMFs are obtained by analogy, which are denoted as c 3 (t)……c i (t), until r i (t) is a monotonic function and the IMF can no longer be separated.

[0148] This application uses the EMD method to decompose the measured signal, which can avoid tedious signal processing, speed up the calculation of fatigue damage of mechanical structures, and improve the efficiency of fatigue analysis.

[0149] The first screening module 530 is used to construct a random white noise sample, establish a logarithmic confidence interval of the energy density of the white noise IMF based on the statistical characteristics of the white noise, and perform a preliminary screening of the IMF components of the load signal to remove false components;

[0150] As an optional implementation, the first screening module 530 is specifically configured to perform the following steps:

[0151] b1: Performing Fourier transform on each order IMF component decomposed by the decomposition module 520 to obtain the frequency spectrum of each order IMF component;

[0152] b2: Calculate the energy density logarithm lnE of each order IMF component i And the average period logarithm value The calculation formulas are:

[0153] oeLh i =ln|C i (f)| 2 (4)

[0154]

[0155] In the formula, C i (f) is the IMF component c i The Fourier transform spectrum of (t), S lnT',i c i (t) Fourier transform spectrum of lnT', T' is the IMF component c i (t) period of Fourier transform;

[0156] b3: Construct several normalized random white noise samples with the same length as the original load signal x(t);

[0157] b4: Perform EMD decomposition on the white noise to obtain the IMF components of each order of white noise. According to the statistical characteristics of white noise, establish the energy density logarithm lnE of each order of white noise IMF components at a specific confidence level j The distribution interval template, lnE j The confidence interval calculation formula is:

[0158]

[0159] In the formula, is the average period logarithm of the Fourier spectrum of each order white noise IMF component, Z α / 2You can check the standard normal distribution table, N is the number of white noise samples, that is, the number of time series, and j represents different white noise IMF components;

[0160] In a specific embodiment, the specific confidence level may be a confidence level of 99%, or may be other specific values ​​set according to circumstances.

[0161] b5: The energy density logarithm of each IMF component decomposed from the original load signal x(t) lnE i and the average period logarithm Compare with the confidence interval established in step b4 to screen out the true components and eliminate useless false components.

[0162] The screening criteria used in this step are that the IMF components outside the confidence interval are valid, and those within the confidence interval are false components. The schematic diagram is shown in Figure 4 As shown, in Figure 4 In the above equation, IMF1, IMF2, IMF4 and IMF5 are valid IMF components, while IMF3, IMF6 and IMF7 are useless false components.

[0163] The second screening module 540 is used to further screen the IMF components initially screened out to screen out the IMF components whose contribution to the total energy value of the original load signal is greater than a predetermined value;

[0164] As an optional implementation, the second screening module 540 is specifically configured to perform the following steps:

[0165] c1: respectively obtain the energy value of each order IMF component screened by the first screening module and the energy value E of the original load signal x , the calculation formulas are:

[0166]

[0167]

[0168] c2: Calculate the contribution β of each order IMF component screened by the first screening module to the total energy value of the original load signal x(t) i , the calculation formula is

[0169]

[0170] c3: According to the contribution calculated in step c2, select the IMF components whose contribution to the total energy value of the original load signal x(t) is greater than a predetermined value.

[0171] In a specific embodiment, the predetermined value may be 0.1, or other appropriate values, which may be determined according to actual conditions. The IMF component screened out in this step is used as the input load for subsequent fatigue damage calculation.

[0172] A first calculation module 550, configured to use the IMF components of the load signal screened by the second screening module as input loads for fatigue analysis, and to respectively calculate fatigue damage caused by each order of IMF components to the mechanical structure;

[0173] In this step, the damage caused by each order IMF component to the mechanical structure is calculated separately, denoted as D 1 , D 2 , D 3 …D i …D n The calculation of the damage caused by each order IMF component to the mechanical structure can be directly performed in finite element analysis software, such as ncode software, etc. The damage calculation method of the mechanical structure can adopt quasi-static method, psd method (self-spectrum method), cpsd method (cross-spectrum method) and modal superposition method, etc. The specific method to be adopted depends on the situation.

[0174] The second calculation module 560 is used to accumulate the fatigue damage caused to the mechanical structure by each IMF component obtained by the first calculation module to obtain the total fatigue damage of the mechanical structure under the action of the original load signal.

[0175] The damage caused by each order IMF component calculated by the first calculation module 550 to the mechanical structure is linearly accumulated to obtain the total fatigue damage D of the mechanical structure under the original excitation or original load signal x(t), that is,

[0176]

[0177] In a third aspect, an embodiment of the present application further provides a terminal, including: a processor, a memory, and a communication unit;

[0178] The memory stores machine-readable instructions executable by the processor, and when the device is running, the processor communicates with the memory through the communication unit;

[0179] The processor executes the machine-readable instructions to perform the methods described in the above aspects.

[0180] The memory can be used to store the execution instructions of the processor, and the memory can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory are executed by the processor, the device is enabled to perform some or all of the steps in the following method embodiments.

[0181] The processor is the control center of the storage terminal, which uses various interfaces and lines to connect various parts of the entire electronic terminal, and executes various functions and / or processes data of the electronic terminal by running or executing software programs and / or modules stored in the memory, and calling data stored in the memory. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor can include only a central processing unit (CPU). In the embodiment of the present application, the CPU can be a single computing core or multiple computing cores.

[0182] The communication unit is used to establish a communication channel so that the storage device can communicate with other terminals, receive user data sent by other terminals or send user data to other terminals.

[0183] In a fourth aspect, the embodiments of the present application further provide a computer storage medium, wherein the computer storage medium may store a program, and when the program is executed, the program may include some or all of the steps in each embodiment provided in the present application. The storage medium may be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0184] The present application adopts EMD decomposition on the original load signal, and screens the IMF based on the statistical characteristics of white noise and the energy contribution, which can avoid the energy leakage and waveform distortion of the load signal after the fast Fourier transform, well preserve the amplitude and frequency of the useful signal, and improve the recognition ability of the subtle features of the original load signal.

[0185] In the embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the node embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0186] The modules described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0187] In addition, each functional module in the embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0188] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. Time domain direct fatigue analysis method based on EMD, It is characterized in that include: Step S1: collecting the load signal x(t) exerted on the mechanical structure; Step S2: performing empirical mode decomposition (EMD) on the collected load signal x(t) to obtain the characteristic mode function (IMF) components of the load signal; Step S3: construct a random white noise sample, establish a logarithmic confidence interval of the energy density of the white noise IMF based on the statistical characteristics of the white noise, perform a preliminary screening of the IMF components of the load signal, and remove false components; Step S4: further screening the IMF components initially screened out to screen out the IMF components whose contribution to the total energy value of the original load signal is greater than a predetermined value; Step S5: using the IMF components of the load signal screened out in step S4 as input loads for fatigue analysis, and calculating fatigue damage caused by each order of IMF components to the mechanical structure; Step S6: Accumulate the fatigue damage caused by each IMF component obtained in step S5 to the mechanical structure to obtain the total fatigue damage of the mechanical structure under the action of the original load signal.

2. The time domain direct fatigue analysis method based on EMD according to claim 1, It is characterized in that The step S2 further comprises: 2-1: Determine all the extreme points of the collected original load signal x(t), and use cubic spline function interpolation to form the upper envelope curve u of the data for all the maximum and minimum points 1 (t) and the lower envelope curve u 2 (t), and take u 1 (t) and u 2 The average value of (t) 1 (t), that is m 1 (t)=(u 1 (t)+u 2 (t)) / 2 (1) 2-2: Let h 1 (t) = x(t) - m 1 (t), if h 1 (t) satisfies two conditions of IMF at the same time: ① The number of zero crossings and the number of extreme points are equal or differ by at most 1 in the entire time history; ② The upper envelope defined by the local maximum and the lower envelope defined by the local minimum are locally symmetrical about the time axis, that is, the mean of the upper envelope defined by the local maximum and the lower envelope defined by the local minimum is 0, then h 1 (t) is the first-order IMF. If it is not satisfied, h 1 (t) is regarded as the new x(t), m 11 (t) is the mean of its upper and lower envelopes, and h 11 (t) = h 1 (t)-m 11 (t), if h 11 (t) is still not satisfied, repeat the process k times, and get h 1k (t)=h 1(k-1) (t)-m 1k (t) (2) If h 1k (t) and h 1(k-1) If the standard deviation SD of (t) is within the predetermined range, the process is stopped. 1k (t) is the first-order IMF of the original load signal x(t), denoted by c 1 (t) = h 1k (t), where the standard deviation SD is calculated as: Where T is the total time length of the load signal x(t); 2-3: Let r 1 (t) = x(t) - c 1 (t), r 1 (t) is regarded as the new x(t), and step 2-2 is repeated to obtain the second-order IMF, denoted as c 2 (t), and then the other order IMFs are obtained by analogy, which are denoted as c 3 (t)……c i (t), until r i (t) is a monotonic function and the IMF can no longer be separated.

3. The time domain direct fatigue analysis method based on EMD according to claim 2, It is characterized in that The step S3 further comprises: 3-1: Perform Fourier transform on each order IMF component obtained in step S2 to obtain the frequency spectrum of each order IMF component; 3-2: Calculate the energy density logarithm lnE of each order IMF component i And the average period logarithm value The calculation formulas are: lnE i =ln|C i (f)| 2 (4) In the formula, C i (f) is the IMF component c i The Fourier transform spectrum of (t), S lnT',i c i (t) Fourier transform spectrum of lnT', T' is the IMF component c i (t) period of Fourier transform; 3-3: Construct several normalized random white noise samples with the same length as the original load signal x(t); 3-4: Perform EMD decomposition on the white noise to obtain IMF components of each order of white noise, and establish the energy density logarithm lnE of each order of white noise IMF components at a specific confidence level according to the statistical characteristics of white noise j The distribution interval template, lnE j The confidence interval calculation formula is: In the formula, is the average period logarithm of the Fourier spectrum of each order white noise IMF component, Z α / 2 You can check the standard normal distribution table, N is the number of white noise samples, that is, the number of time series, and j represents different white noise IMF components; 3-5: The energy density logarithm of each IMF component decomposed from the original load signal x(t) is lnE i and the average period logarithm Compare with the confidence interval established in step 3-4 to filter out the true components and eliminate useless false components.

4. The time domain direct fatigue analysis method based on EMD according to claim 3, It is characterized in that The step S4 further comprises: 4-1: Calculate the energy value of each order IMF component screened in step S3 and the energy value E of the original load signal x , the calculation formulas are: 4-2: Calculate the contribution β of each order IMF component screened in step S3 to the total energy value of the original load signal x(t) i , the calculation formula is 4-3: According to the contribution calculated in step 4-2, the IMF components whose contribution to the total energy value of the original load signal x(t) is greater than a predetermined value are screened out.

5. Time domain direct fatigue analysis device based on EMD, It is characterized in that include: An acquisition module, used for acquiring a load signal x(t) received by the mechanical structure; A decomposition module is used to perform empirical mode decomposition (EMD) on the collected load signal x(t) to obtain each characteristic mode function (IMF) component of the load signal; The first screening module is used to construct a random white noise sample, establish a logarithmic confidence interval of the energy density of the white noise IMF based on the statistical characteristics of the white noise, perform a preliminary screening of the IMF components of the load signal, and remove false components; A second screening module is used to further screen the IMF components initially screened out to screen out the IMF components whose contribution to the total energy value of the original load signal is greater than a predetermined value; A first calculation module, used for taking the IMF components of the load signal screened by the second screening module as input loads for fatigue analysis, and respectively calculating fatigue damage caused by each order of IMF components to the mechanical structure; The second calculation module is used to accumulate the fatigue damage caused to the mechanical structure by each IMF component obtained by the first calculation module to obtain the total fatigue damage of the mechanical structure under the action of the original load signal.

6. The time domain direct fatigue analysis device based on EMD as claimed in claim 5, It is characterized in that The decomposition module is specifically used to perform the following steps: a1: Determine all the extreme points of the collected original load signal x(t), and use cubic spline function interpolation to form the upper envelope curve u of the data for all the maximum and minimum points 1 (t) and the lower envelope curve u 2 (t), and take u 1 (t) and u 2 The average value of (t) 1 (t), that is m 1 (t)=(u 1 (t)+u 2 (t)) / 2 (1) a2: Let h 1 (t) = x(t) - m 1 (t), if h 1 (t) satisfies two conditions of IMF at the same time: ① The number of zero crossings and the number of extreme points are equal or differ by at most 1 in the entire time history; ② The upper envelope defined by the local maximum and the lower envelope defined by the local minimum are locally symmetrical about the time axis, that is, the mean of the upper envelope defined by the local maximum and the lower envelope defined by the local minimum is 0, then h 1 (t) is the first-order IMF. If it is not satisfied, h 1 (t) is regarded as the new x(t), m 11 (t) is the mean of its upper and lower envelopes, and h 11 (t) = h 1 (t)-m 11 (t), if h 11 (t) is still not satisfied, repeat the process k times, and get h 1k (t)=h 1(k-1) (t)-m 1k (t) (2) If h 1k (t) and h 1(k-1) If the standard deviation SD of (t) is within the predetermined range, the process is stopped. 1k (t) is the first-order IMF of the original load signal x(t), denoted by c 1 (t) = h 1k (t), where the standard deviation SD is calculated as: Where T is the total time length of the load signal x(t); a3: Let r 1 (t) = x(t) - c 1 (t), r 1 (t) is regarded as the new x(t), and step a2 is repeated to obtain the second-order IMF, denoted as c 2 (t), and then the other order IMFs are obtained by analogy, which are denoted as c 3 (t)……c i (t), until r i (t) is a monotonic function and the IMF can no longer be separated.

7. The time domain direct fatigue analysis device based on EMD as claimed in claim 6, It is characterized in that The first screening module is specifically used to perform the following steps: b1: Perform Fourier transform on each order IMF component obtained by the decomposition module to obtain the spectrum of each order IMF component; b2: Calculate the energy density logarithm lnE of each order IMF component i And the average period logarithm value The calculation formulas are: lnE i =ln|C i (f)| 2 (4) In the formula, C i (f) is the IMF component c i The Fourier transform spectrum of (t), S lnT',i c i (t) Fourier transform spectrum of lnT', T' is the IMF component c i (t) period of Fourier transform; b3: Construct several normalized random white noise samples with the same length as the original load signal x(t); b4: Perform EMD decomposition on the white noise to obtain the IMF components of each order of white noise. According to the statistical characteristics of white noise, establish the energy density logarithm lnE of each order of white noise IMF components at a specific confidence level j The distribution interval template, lnE j The confidence interval calculation formula is: In the formula, is the average period logarithm of the Fourier spectrum of each order white noise IMF component, Z α / 2 You can check the standard normal distribution table, N is the number of white noise samples, that is, the number of time series, and j represents different white noise IMF components; b5: The energy density logarithm of each IMF component decomposed from the original load signal x(t) is lnE i and the average period logarithm Compare with the confidence interval established in step b4 to screen out the true components and eliminate useless false components.

8. The time domain direct fatigue analysis device based on EMD as claimed in claim 7, It is characterized in that The second screening module is specifically used to perform the following steps: c1: respectively obtain the energy value of each order IMF component screened by the first screening module and the energy value E of the original load signal x , the calculation formulas are: c2: Calculate the contribution β of each order IMF component screened by the first screening module to the total energy value of the original load signal x(t) i , the calculation formula is c3: According to the contribution calculated in step c2, select the IMF components whose contribution to the total energy value of the original load signal x(t) is greater than a predetermined value.

9. A terminal, It is characterized in that include: Processor, memory and communication unit; The memory stores machine-readable instructions executable by the processor, and when the device is running, the processor communicates with the memory through the communication unit; The processor executes the machine-readable instructions to perform the method according to any one of claims 1 to 4.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is executed.

Images