A method for compiling a measured severe wind spectrum based on a Monte Carlo method

Abstract

The application discloses a kind of based on monte carlo method's measured severe gust spectrum compilation method, comprising: S1. clear aircraft mission profile;S2. load data preparation;S3. load data statistical processing;S4. obtain severe gust overload cumulative exceedance curve: monte carlo method is simulated group mission profile gust load time history, and then the damage distribution of each mission profile is calculated, and the reliability P L 90% of the damage as severe damage;Further based on severe damage, the coverage rate P t of corresponding cumulative exceedance under specified overload is obtained;The cumulative exceedance of the coverage rate P t under each level overload is taken, and the severe overload cumulative exceedance curve under each mission section is fitted;S5. determine 5 levels of load by load spectrum equivalent criterion, and compile severe gust overload spectrum;The application can reduce the time of fatigue test, and the dispersion coefficient used in life determination only needs to consider the dispersion coefficient of structure.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of flight load spectrum, more particularly to a measured severe gust spectrum compilation method based on Monte Carlo method. BACKGROUND

[0002] Flight load spectrum refers to a spectrum of load time history of an aircraft body in flight. The measured flight load spectrum of an aircraft is compiled through special test modification and flight test, and is used to determine and verify the design service life of the aircraft, which is a prerequisite for the fatigue life extension of the aircraft structure. Gust load accounts for a high proportion of damage in flight load, and the compilation of the measured gust spectrum of the aircraft is of great significance to the life extension of military / civil aircraft. The aviation industry has accumulated a large amount of measurement data, and formed a gust spectrum compilation method based on gust speed exceedance number curve, including the 'TWIST' method of Europe and the '5x5' spectrum method of Boeing. However, these methods are generally used to compile the average intensity spectrum of gust, and cannot reflect the dispersion degree of gust.

[0003] Even if the aircrafts used under the same usage requirements, the gust load-time history of different aircrafts in the fleet has obvious differences, corresponding to the dispersion of the load spectrum, and the durability analysis and test must be carried out under the determined (unique) load spectrum. Therefore, how to select and compile a reasonable load spectrum becomes the key to structural durability analysis and evaluation. For this reason, the concept of "severe spectrum" is proposed, which has the advantages of exposing the failure characteristics of the aircraft itself and reducing the test time, so the application of severe spectrum in the compilation of the measured spectrum is also increasing.

[0004] Therefore, how to propose a measured severe gust spectrum compilation method of the aircraft which fully considers the gust dispersion under the condition of ensuring safety and economy is a problem to be solved by those skilled in the art. SUMMARY

[0005] In view of the above, the present application provides a measured severe gust spectrum compilation method based on Monte Carlo method

[0006] In order to achieve the above purpose, the present application adopts the following technical scheme:

[0007] A measured severe gust spectrum compilation method based on Monte Carlo method, comprising the following steps:

[0008] S1. Defining the mission profile of the aircraft: according to the usage requirements of the aircraft, the composition and proportion of the mission profile are given, and for each mission profile, the mission profile parameters are defined;

[0009] S2. Load data preparation: pre-processing the measured load data obtained according to the mission profile to obtain load data;

[0010] S3. Statistical processing of load data: estimating the distribution parameters of the exceedances under the specified overload according to the maximum likelihood estimation function;

[0011] S4. Calculating the damage distribution of each mission profile, taking the reliability P of the fleet L of 90% as the severe damage; further based on the severe damage, the severe overload cumulative exceedance curve under each mission segment is obtained by inversion:

[0012] S41. Based on the distribution parameters of the exceedances under the specified overload, the Monte Carlo method is used to randomly extract the gust overload versus time sequence (n y,max , n y,min ) i ; where i is the level of overload, j is each mission segment;

[0013] S42. According to the gust overload versus time sequence (n y,max , n y,min ) i , the single-cycle equivalent damage is calculated, and the load spectrum damage D eq under single landing is obtained by accumulation;

[0014] S43. According to the Weibull distribution of load spectrum damage D eq , the Weibull distribution parameters α and β are calculated by fitting D eqi and the corresponding empirical frequency value f i ; where D eqi is the load spectrum damage corresponding to the i-th sample data;

[0015] S44. Set the load spectrum reliability P L to 90%, and calculate the specific method of exceedance coverage P t :

[0016] D eq90 = exp((ln(-ln(1-0.9))+αlnβ) / α)

[0017] Inversion to obtain the corresponding u eqpl and the corresponding exceedance coverage P pt of the current severe damage D t , where u pt is the quantile of the standard normal distribution corresponding to the exceedance coverage P t , which should satisfy that the current severe damage D eq90 is equal to the damage corresponding to the overload cumulative exceedance curve with exceedance coverage P t :

[0018]

[0019] Where μ i, σ i respectively, the i-th level overload n y,i the corresponding logarithmic mean and standard deviation of the transfinite number u Pt is the coverage rate P t the corresponding standard normal distribution quantile;

[0020] The following equation is solved by dichotomy:

[0021]

[0022] u Pt satisfies the requirement, and the corresponding transfinite number coverage rate P t is obtained.

[0023] The transfinite number coverage rate P tk of each task profile is obtained by the above method, wherein k represents the serial number of the task profile;

[0024] For each level load of each task segment, the transfinite number with a coverage rate P t is calculated according to the distribution function and the distribution parameter in S3, that is

[0025]

[0026] wherein μ i , σ i respectively, the i-th level overload n y,i the corresponding logarithmic mean and standard deviation of the transfinite number u

[0027] After obtaining the data pair of (Δn y , ΔN Pt ) i , the corresponding cumulative transfinite number curve of the severe gust overload is obtained by fitting;

[0028] S5. The 5th level load is determined by the load spectrum equivalent criterion, and the severe gust overload spectrum is compiled by the root 5x5 spectrum: for the typical task segment in the task profile, the severe task segment spectrum is compiled; the severe task profile spectrum (take-off and landing spectrum) is composed of the task segment; the fly-continuous-fly spectrum with 1000 take-offs and landings as a cycle is compiled by randomly sorting the task profile severe spectrum in proportion.

[0029] Preferably, the specific content of S1 includes:

[0030] The task profile, the task profile proportion and the task profile composition of the aircraft are determined, and the task profile parameters are given, wherein the task profile parameters include: task segment, height, speed, weight, flight distance and flight time of the task segment; and the division of the measured data task segment, and the division principle is:

[0031] a) The criterion for the start of the climb is that the flap deflection angle becomes 0°;

[0032] b) The criterion for the end of the climb / the start of the level flight is the turning point of the height curve;

[0033] c) The criterion for the end of the level flight / the start of the glide is the turning point of the height curve;

[0034] d) The criterion for the end of the glide is that the flap deflection angle becomes 35°.

[0035] Preferably, the specific contents of the pre-processing of the measured load data in S2 include:

[0036] (1) Standardization of the overload data n y

[0037] △n y0 =△n yi *Gi / G0

[0038] In the formula,△n y0 is the overload value after correction according to the standard mission segment;△n yi is the measured overload value, G i is the actual mass of the aircraft corresponding to the measured overload value, and G0 is the standard aircraft mass of the flight profile in the mission segment;

[0039] (2) Separation of gust and maneuver overloads

[0040] The maneuver and gust loads contained in the measured load data are separated, and the interpretation conditions of the maneuver and gust load spectrum of the hollow flight are:

[0041] a) The overload that changes slowly and lasts more than 2s is a maneuver, otherwise it is a gust;

[0042] b) The overload with a rudder deflection angle absolute value equal to or greater than 4° is a maneuver overload, otherwise it is a gust overload;

[0043] c) For the climb and glide mission segments, since there are fewer maneuver actions, the loads are approximately considered as gust loads, and the separation of gust and maneuver is not performed;

[0044] (3) Peak and valley value collection

[0045] In the counting process, all peak and valley values are detected and represented in order as Δn zi , the data between the peak and valley values are filtered out, while the corresponding sampling point sequence numbers are retained, if:

[0046] or

[0047] the peak or valley value is taken once.​

[0048] Preferably, S3 also includes:

[0049] (1) The overload time history counting adopts the limited cross-peak counting method;

[0050] Using the 1g load state of each task segment as a benchmark, when the valley value between two peaks or the peak value between two valleys does not exceed the deviation of the benchmark line for that task segment:

[0051] For positive overload, the deviation is the upper deviation, and only the maximum peak value is recorded;

[0052] For negative overload, the deviation is the lower deviation, and only the minimum valley value is recorded;

[0053] The absolute values ​​of both the upper and lower deviations are 20% of the maximum peak value;

[0054] (2) Obtain the cumulative exceedance number of each load level;

[0055] The peak value obtained by the cross-peak counting method is defined as a positive peak value and a valley value as a negative valley value. Several load levels are selected, and the cumulative exceedance counts of positive and negative gust overloads are recorded for both positive and negative peak values, denoted as N(+Δn). y ) and N(-Δn y );

[0056] Get + Δn y Cumulative transcendence number:

[0057]

[0058] (3) The overload cumulative exceedance number is standardized considering the flight duration;

[0059] The measured data corresponds to the flight time t. M The standard time for the task segment is t. S The cumulative exceedance number N of the i-th level load obtained from the measured time. ib Then N ib The standardization rules are as follows:

[0060]

[0061] Where N i The cumulative exceedance number of the i-th level load over standard time;

[0062] The cumulative overload exceedances of each load level are standardized to obtain the cumulative overload exceedance data for the task segment at standard time.

[0063] Preferably, the method for estimating the distribution parameters of the exceedance number under the specified overload in S3 is as follows:

[0064] The maximum likelihood estimation method is used to estimate the distribution parameters; the likelihood function is:

[0065]

[0066] In the formula x i =lgΔN i ;

[0067] The objective parameters (μ, σ) are solved by finding that the derivatives of the likelihood functions with respect to μ and σ are zero.

[0068]

[0069] Preferably, the specific content of S41 includes:

[0070] 1) Randomly select a random number u that follows a standard normal distribution. p u p The quantile of the standard normal distribution corresponding to the coverage rate P, and the i-th level gust overload Δn yi The corresponding coverage is the transcendence number ΔN of P. P i is:

[0071]

[0072] 2) For ΔN i Cumulative counting yields the cumulative exceedance number. Therefore, from (Δn) y ,N) i The data was fitted to obtain the Δn of the task profile. y Transcendental number curve;

[0073] 3) Repeat step 1) for all task segments to obtain the overload exceedance curves of all task segments under the task profile represented by this sampling.

[0074] 4) Discretize the family of measured cumulative exceedance number curves for sudden gust overload to obtain exceedance number data pairs (±Δn) corresponding to each level of sudden gust overload. y* ,ΔN) ij Where i represents the overload level and j represents each task segment; the discretization process is as follows: the transcendence number ΔN is calculated. i =N i -N i-1 Obtain exceedance number data pairs (Δn) corresponding to different sudden wind overloads. y ,ΔN) i , i = 1, ..., n;

[0075] 5) Set the specified Δn for each task segment under the task segment where the sudden wind overload occurs. yThe exceedance numbers are summed to obtain the total exceedance number data for each level of sudden wind overload. m is the number of task segments;

[0076] 6) For sudden gust overload, sort the positive and negative overload levels according to their absolute values ​​and match them to obtain the time series of sudden gust overload pairs for this task profile under this sampling (n y,max n y,min ) i ;

[0077] 7) Repeat the above sampling process for each task profile. For each task profile, a family of time series of sudden wind overload pairs is obtained.

[0078] Preferably, in step 2), the Δn of a single takeoff and landing is obtained by fitting. y The specific method for creating transcendental number curves is as follows:

[0079] For each task segment, the cumulative overload exceedance data pair (Δn) y ,N) i The fitting equation is as follows:

[0080] Δn y =a*lgN+b

[0081] The overload cumulative exceedance curve for each task segment was calculated.

[0082] Preferably, the specific content of S42 includes:

[0083] Based on the compiled time series of sudden wind overload (n) y,max n y,min ) i The equivalent damage for a single cycle is calculated, and the cumulative load spectrum damage D of the mission profile is obtained. eq The specific method is as follows:

[0084]

[0085] In the formula, m is the damage index, i.e., the slope of the material / structure fatigue SN curve when the stress ratio R = 0, and d eq This refers to damage in a single cycle.

[0086] Preferably, S4 also includes: S45. Performing high-load truncation and low-load truncation on the severe cumulative exceedance curve of the task segment;

[0087] (1) High load interception

[0088] Determine the level and number of high loads in the severe gale spectrum, and select high loads at a preset frequency as the cutoff value;

[0089] (2) Low-load cutoff

[0090] The small amplitude load in the severe gust spectrum is deleted or converted into a certain level load.

[0091] Compared with the prior art, the method for compiling a measured severe gust spectrum based on a Monte Carlo method is provided, which comprises the following steps: determining the severe damage reliability PL of a fleet; simulating the gust load-time distribution of the fleet by using a Monte Carlo method; calculating the damage distribution of the fleet; taking the damage of the fleet with the reliability PL as the severe damage; obtaining the coverage rate Pt of the cumulative exceedance number under a specified overload based on the inversion of the damage; taking the cumulative exceedance number under the coverage rate Pt of each level overload; fitting the severe overload cumulative exceedance number curve; and compiling the severe gust spectrum by using the curve. BRIEF DESCRIPTION OF DRAWINGS

[0092] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only constitute the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on the provided drawings.

[0093] Figure 1 A flowchart of the method for compiling a measured severe gust spectrum based on a Monte Carlo method provided by the present application;

[0094] Figure 2 A peak-valley value detection schematic diagram in the method for compiling a measured severe gust spectrum based on a Monte Carlo method provided by the present application;

[0095] Figure 3 A limit cross-peak value counting schematic diagram in the method for compiling a measured severe gust spectrum based on a Monte Carlo method provided by the present application;

[0096] Figure 4 A μ fitting result schematic diagram in the distribution parameter estimation of the method for compiling a measured severe gust spectrum based on a Monte Carlo method provided by the present application;

[0097] Figure 5 A σ fitting result schematic diagram in the distribution parameter estimation of the method for compiling a measured severe gust spectrum based on a Monte Carlo method provided by the present application;

[0098] Figure 6 A fleet damage distribution schematic diagram in the method for compiling a measured severe gust spectrum based on a Monte Carlo method provided by the present application;

[0099] Figure 7 A schematic diagram of a severe gust overload cumulative exceedance number curve in a measured severe gust spectrum compilation method based on a Monte Carlo method provided by the present application is shown in the figure;

[0100] Figure 8 A flight type random spectrum in a measured severe gust spectrum compilation method based on a Monte Carlo method provided by the present application is shown in the figure; DETAILED DESCRIPTION

[0101] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present application.

[0102] The embodiments of the present application disclose a measured severe gust spectrum compilation method based on a Monte Carlo method, as shown in the figure, comprising the following steps: Figure 1

[0103] S1. Defining an aircraft mission profile: according to the use requirements of the aircraft, the composition and proportion of the mission profile are given, and for each mission profile, the mission profile parameters are defined;

[0104] S2. Load data preparation: the measured load data obtained according to the mission profile is preprocessed to obtain load data;

[0105] S3. Statistical processing of load data: the distribution parameters of the exceedance number under a specified overload are estimated according to the maximum likelihood estimation function;

[0106] S4. Calculating the damage distribution of each mission profile, taking the reliability P L of the fleet as 90% of the damage as severe damage; and then based on the severe damage, the severe overload cumulative exceedance number curve under each mission segment is obtained by inversion:

[0107] S41. Based on the distribution parameters of the exceedance number under a specified overload, the Monte Carlo method is used to randomly extract the gust overload versus time sequence (n y,max , n y,min ) i ; wherein i is the level of the overload, and j is each mission segment;

[0108] S42. According to the gust overload versus time sequence (n y,max , n y,min ) i , the single-cycle equivalent damage is calculated, and the load spectrum damage D eq under the mission profile is obtained by accumulation; ​

[0109] S43. The load spectrum damage D under each mission profile is damaged eq All obey Weibull distribution, fitting D eqi And the corresponding empirical frequency value f i The Weibull distribution parameters a and b of the load spectrum damage of each mission profile are calculated; wherein D eqi is the load spectrum damage corresponding to the ith sample data;

[0110] S44. Set the load spectrum reliability P L to 90%, calculate the specific method of the exceedance coverage P t :

[0111] D eq90 = exp((ln(-ln(1-0.9))+a ln b) / a)

[0112] The current severe damage D eq90 corresponding to u pt and the corresponding exceedance coverage P t is obtained by inversion, wherein u pt is the quantile of the standard normal distribution corresponding to the exceedance coverage P t , which should satisfy the current severe damage D eq90 equal to the damage corresponding to the overload cumulative exceedance curve with exceedance coverage P t :

[0113]

[0114] Wherein, μ i , σ i are the logarithmic median and standard deviation of the exceedance number corresponding to the nth overload n y,i , u Pt is the quantile of the standard normal distribution corresponding to the coverage P t ;

[0115] The following equation is solved by dichotomy:

[0116]

[0117] Get the required u Pt , which should be accurate to the third decimal place, and the corresponding exceedance coverage P t is obtained.

[0118] The exceedance coverage P tk of each mission profile is obtained by the above method, wherein k represents the serial number of the mission profile.

[0119] For each level load of each task section under each task profile, the coverage rate P is calculated according to the distribution function and distribution parameters in S3 t The hyper number of the hyper number is

[0120]

[0121] Wherein, μ i , σ i are the logarithmic mean and standard deviation of the hyper number of the corresponding hyper number of the ith level overload n y,i .

[0122] After obtaining the data pair of (Δn y , ΔN Pt ) i , the corresponding hyper number curve of the severe gust overload is obtained by fitting.

[0123] S5. The 5th level load is determined by the load spectrum equivalent criterion, and the severe gust overload spectrum is compiled by the root 5x5 spectrum: for the typical task section in the task profile, the severe task section spectrum is compiled; the severe spectrum of the task profile is composed of the task section (take-off and landing spectrum); the fly-continue-fly spectrum with 1000 take-off and landings as a cycle is compiled by randomly sorting the severe spectrum of the task profile in proportion.

[0124] It should be noted that: the severe spectrum should meet the requirements of GJB 67.6A-2008 Military Aircraft Strength Specification: Repeated Load, Durability and Damage Tolerance about the durability severe spectrum, according to the requirements of the standard, the durability severe spectrum should reflect the 90% usage of the fleet. The Monte Carlo method is used to simulate the gust load-time distribution of the fleet in the present application, and then the damage distribution of the fleet is calculated, and the damage of the fleet with a reliability of 90% is taken as the severe damage. Further based on the inversion of the damage, the coverage rate Pt of the cumulative hyper number under the corresponding specified overload is obtained, and the cumulative hyper number with the coverage rate Pt under each level overload is taken to fit the severe overload cumulative hyper number curve.

[0125] In S43, it is assumed that the damage D eq obeys the Weibull distribution, and the distribution parameters are calculated by using the rank statistics method, as shown in the following formula, and the Weibull distribution parameters α and β are calculated by fitting D eqi and the corresponding empirical frequency value f i .

[0126] ln(-ln(1-f i ))=αlnD eqi -αlnβ

[0127] In order to further implement the above technical scheme, the specific content of S1 includes:

[0128] Determine the mission profile, mission profile ratio and mission profile composition of the aircraft, and give the mission profile parameters, wherein the mission profile parameters include: mission segment, height, speed, weight, flight distance and flight time of the mission segment; and the division of the measured data mission segment, the division principle is:

[0129] a) The climb start point is determined by the flap angle changing to 0°;

[0130] b) The climb end point / flat start point is determined by the turning point of the height curve;

[0131] c) The flat end point / gliding start point is determined by the turning point of the height curve;

[0132] d) The glide end point is determined by the flap angle changing to 35°.

[0133] It should be noted that the overload spectrum of the main mission segment of the hollow flight profile is compiled according to the "5x5" spectrum, that is, the overload spectrum is first discretized into five levels of spectrum, and then the discretized spectrum is compiled into load spectrum of five typical flight types with different degrees of severity. There are five typical flight types.

[0134] In order to further implement the above technical scheme, the specific content of the pre-processing of the measured load data in S2 includes:

[0135] (1) Standardization processing of overload data n y

[0136] △n y0 =△n yi *Gi / G0

[0137] In the formula, △n y0 is the overload value corrected according to the standard mission segment; △n yi is the measured overload value, G i is the actual mass of the aircraft corresponding to the measured overload value, and G0 is the standard aircraft mass of the flight profile at this mission segment; the measured center of gravity overload n y minus 1 can obtain the incremental overload Δn y at this time.

[0138] It should be noted that the load actual mass data is given by subtracting the fuel consumption from the weight of the aircraft, and the fuel consumption is obtained by total average calculation (the average fuel consumption calculation time is from the start of take-off taxiing to the end of landing impact).

[0139] (2) Separation of gust and maneuver overload

[0140] The maneuver and gust load contained in the measured load data is separated, and the judgment condition of the maneuver and gust load spectrum of the hollow flight is:

[0141] ​a) Overload changes slowly and lasts more than 2s, it is maneuver, otherwise it is gust;

[0142] b) Overload with rudder deflection absolute value equal to or greater than 4° is maneuver overload, otherwise it is gust overload;

[0143] c) For climb and glide mission segments, due to less maneuver action, the loads are approximately considered as gust loads, and no longer separate gust and maneuver;

[0144] (3) Peak and valley value collection

[0145] In the counting process, all peak and valley values are detected and expressed in sequence as Δn zi , the data between peak and valley values are filtered out, while the corresponding sampling point sequence numbers are retained, if it meets:

[0146] or

[0147] then take the peak or valley value once.

[0148] In this fact example, as shown in Figure 2 , the peak and valley points of Δn z1 , Δn z3 , Δn z5 , Δn z6 , Δn z7 , Δn z8 , Δn z10 are retained, and the sampling points of Δn z2 , Δn z4 , Δn z9 are removed.

[0149] In order to further implement the above technical solution, S3 further includes:

[0150] (1) The overload time history counting adopts the limited cross-peak counting method;

[0151] The limited cross-peak counting method is a counting method formed by adding certain limitation conditions on the basis of the peak method. As shown in Figure 3 , the specific processing requirements are as follows: taking the 1g load state of each mission segment as a reference, when the peak value between two peaks or the valley value between two valleys does not exceed the deviation of the reference line of the mission segment:

[0152] For positive overload, the deviation is the upper deviation, and only the maximum peak value is recorded;

[0153] For negative overload, the deviation is the lower deviation, and only the minimum valley value is recorded;

[0154] The absolute values of the upper deviation and the lower deviation are both 20% of the maximum peak value;

[0155] (2) Obtain the cumulative number of overruns of each level of load;

[0156] The peak value greater than 0 obtained by the limit cross-peak counting method is a positive peak value, and the valley value less than 0 is a negative valley value; a plurality of levels of loads are selected, and the positive peak value and the negative valley value are recorded as the positive gust overload cumulative overrun number and the negative gust overload cumulative overrun number, respectively, and are recorded as N(+Δn y ) and N(-Δn y ) respectively;

[0157] The discrete gust model considers that for the same intensity (i.e. the same amplitude) of the gust disturbance, the occurrence probability of the positive direction (upward) and the negative direction (downward) is equal, so theoretically the positive and negative gust overload cumulative overrun number curves obtained by measurement should be symmetrical. However, in the actual process, there is a certain difference between the positive and negative gust cumulative overrun number curves, and the geometric mean is used in engineering to obtain an average of the cumulative overrun number corresponding to the symmetrical gust overload. Obtain the cumulative overrun number of Δn + Δn y

[0158]

[0159] (3) Consider the overload cumulative overrun number standardization processing of flight duration;

[0160] Since the load spectrum measurement is different from the task profile standard time of the task segment flight time, it is necessary to convert the measured task segment overload cumulative overrun number data to the standard time cumulative overrun number data. The flight time corresponding to the measured data is t M , the task segment standard time is t S , the cumulative overrun number of the i-th level of load obtained by the measured time is N ib , and the standardization rule of N ib is:

[0161]

[0162] Where N i is the cumulative overrun number of the i-th level of load of the standard time;

[0163] The cumulative overrun number of each level of load is standardized to obtain the task segment overload cumulative overrun number data of the standard time.

[0164] It should be noted that: it also includes the overrun number distribution characteristic test: usually a random variable model is used, and it is assumed that the overrun number ΔN corresponding to the specified Δn y obeys a lognormal distribution.

[0165] In order to further implement the above technical scheme, the distribution parameter estimation method in S3 is:

[0166] ​The maximum likelihood estimation method is used to estimate the distribution parameters; the likelihood function is:

[0167]

[0168] where x i = lgΔN i ;

[0169] The target parameters (μ and σ) are solved by taking the derivative of the likelihood function with respect to μ and σ respectively:

[0170]

[0171] To further implement the above technical scheme, the specific content of S41 includes:

[0172] 1) Randomly extract a random number u p , u p is the quantile of the standard normal distribution corresponding to the coverage P, and the ith level of the sudden wind overload Δn yi corresponds to the coverage of the super number ΔN P , i is:

[0173]

[0174] 2) Accumulate the cumulative number of ΔN i , and obtain the cumulative super number Thus, the Δny super number curve of the task profile is fitted from the (Δn y , N) i data pair;

[0175] 3) Repeat step 1) for all task segments to obtain the sudden wind overload super number curve of all task segments represented by the task profile in this sampling;

[0176] 4) Discretize the measured sudden wind overload cumulative super number curve family to obtain the super number data pair (±Δn y* , ΔN) ij corresponding to each level of sudden wind overload, where i is the level of the overload, and j is each task segment; wherein the discretization process is: calculate the super number ΔN i = N i -N i-1 , obtain the super number data pair (Δn y , ΔN) i corresponding to different sudden wind overloads, i = 1, …, n;

[0177] 5) Sum the super numbers of the specified Δn y of each task segment under the sudden wind overload of the task segment, respectively, to obtain the total super number data pair of each level of sudden wind overload m is the number of task segments;

[0178] 6) For sudden gust overload, sort the positive and negative overload levels according to their absolute values ​​and match them to obtain the time series of sudden gust overload pairs for this task profile under this sampling (n y,max n y,min ) i ;

[0179] 7) Repeat the above sampling process for each task profile. For each task profile, a family of time series of sudden wind overload pairs is obtained.

[0180] To further implement the above technical solution, Δn of the task profile is obtained by fitting in step 2). y The specific method for creating transcendental number curves is as follows:

[0181] For each task segment, the cumulative overload exceedance data pair (Δn) y ,N) i The fitting equation is as follows:

[0182] Δn y =a*lgN+b

[0183] The overload cumulative exceedance curve for each task segment was calculated.

[0184] To further implement the above technical solution, the specific content of S42 includes:

[0185] Based on the compiled time series of sudden wind overload (n) y,max n y,min ) i The equivalent damage for a single cycle is calculated, and the cumulative load spectrum damage D is obtained. eq The specific method is as follows:

[0186]

[0187] In the formula, m is the damage index, which is the slope of the material / structure fatigue SN curve when R=0. For aluminum alloy materials, it can be approximated as 4.

[0188] It should be noted that the load damage is calculated using the Odin transform + linear cumulative damage method, and the overload spectrum is used as an example in this embodiment.

[0189] To further implement the above technical solution, S4 also includes: S45. High-load truncation and low-load truncation of the severe cumulative exceedance curve of the task segment;

[0190] (1) High load interception

[0191] Determine the level and number of high loads in the severe gale spectrum, and select high loads at a preset frequency as the cutoff value;

[0192] (2) Low load cut-off

[0193] The small amplitude loads in the severe gust spectrum are deleted or equal damage is converted to a certain level load.

[0194] It should be noted that:

[0195] High load cut-off: The high loads in the load spectrum have two effects: on the one hand, they increase fatigue damage, so they need to be supplemented in the spectrum according to the situation of the load spectrum; on the other hand, the coupling effect of more high loads and low loads increases the life. Therefore, the level and quantity of high loads in the spectrum need to be determined; usually, the high load that occurs once in 1000 flights of the mission segment is taken as the cut-off value.

[0196] Low load cut-off: The load spectrum usually contains a large number of small amplitude load cycles, which need to be deleted or equal damage converted to a certain level load. Usually, the fatigue limit of the key position is determined according to the foregoing analysis, and the overload value corresponding to 70% to 80% of the fatigue limit is cut off.

[0197] The specific process of determining the 5-level load from the load spectrum equivalent criterion and the severe gust overload spectrum compiled by the root 5x5 spectrum will be further described below:

[0198] 1. Level load determination

[0199] (1) Load level determination based on mission profile

[0200] The hollow flight mission segments are discretized into 5 levels. The number of loads after discretization is an integer in a program block, and the number of flights for each mission segment is not less than 1 time, and the representative value of each level load (i.e. equivalent load) is determined by equal damage conversion of the discretized segment.

[0201] (2) 5x5 spectrum level load parameter determination

[0202] Based on the severe gust overload cumulative exceedance curve of the mission segment, the load spectrum equivalent calculation is carried out according to the method of "Civil Aircraft Structure Durability and Damage Tolerance Design Manual Volume 1". The calculation of equivalent load: assuming that a certain discretized segment needs to be calculated, the equivalent load of the discretized segment is Δn yd , the equivalent load cycle number is N eq , and m linear segments are used to replace the curve of the discretized segment, and the linear equation of the i-th segment is:

[0203] Δg = a i lgN + b i

[0204] In the formula, a i , b i are the constants of the i-th segment of the load spectrum curve.

[0205] Then the equivalent load is:

[0206]

[0207] wherein:

[0208]

[0209] S is the slope parameter of the material S-N curve, S = 2.0 for aluminum alloy.

[0210] Equivalent load cycle number

[0211] 2. Determination of flight type

[0212] The load spectrum of each flight mission profile of the hollow flight is compiled according to 5 different flight types respectively. The principle of determining the typical flight type of the hollow flight is: taking the gust spectrum of the flight (glide) segment with the highest load as the benchmark, according to the assumption that the highest load in each flight (glide) segment is the extreme value of the normal logarithmic distribution, the number of occurrence of each flight type is determined (y i is determined); according to the assumption that the gust spectrum shape of each flight type is similar, the gust increment overload spectrum of each flight type in 1000 flights is compiled (B ij is determined); as shown in Table 1.

[0213] Table 1 5x5 spectrum

[0214]

[0215] The y1+y2+y3+y4+y5 in the table should be equal to the number of occurrences of the profile in 1000 flights. The 5x5 spectrum of each mission segment under all profiles is compiled according to the above method. And the y1, y2, y3, y4, y5 of the 5x5 spectrum of each mission segment under each mission profile are the same.

[0216] 3. Serious gust overload spectrum compilation

[0217] (1) Compile mission segment spectrum

[0218] According to the data in the 5x5 spectrum compiled in 4.4.2, the load at each level and its corresponding frequency of the kth flight type of the ith mission segment of the jth profile under single take-off and landing are associated, and the peak and valley values are randomly selected alternately (i.e. the peak and valley values are randomly selected respectively) to randomly pair and arrange, forming the load spectrum of the specific flight type of the mission segment, denoted as (Δn ydn ,-Δn ydm ) h , and the mission segment spectrum (jxixk) is expressed in the form of the following load pair sequence

[0219] f i,j,k =(Δn ydn ,-Δnydm )h

[0220] where Δn ydn is the n(n=1,2,3,4,5) level load of the mission segment, representing the peak load; and Δn ydm is the m(m=1,2,3,4,5) level load of the mission segment, representing the valley load. The total number of load pairs for each flight type of mission segment should be equal to the total number of cycles for each flight in the above table. In this way, the spectrum (load pair sequence) for each flight type of all mission segments is compiled.

[0221] (2) Compile the mission profile spectrum

[0222] Let the load spectrum of the kth flight type of the jth profile be F j,k . F j,k can be expressed as a load pair sequence f i,j,k , which is sorted according to the order of the mission segment i(i=1,2,3...m) under the profile, i.e., f 1,j,k ,f 2,j,k ,f 3,j,k ,f 4,j,k ……f m,j,k . In this way, the mission profile spectrum under a complete flight is obtained. The mission profile spectrum for each flight type of all mission profiles is compiled according to the above method. All kinds of mission profile spectra are expressed in vector form as

[0223] B=[F 1,1 … F 1,K F 2,1 … F 2,K F 3,1 … F 3,K …… F L,1 …F L,K ]

[0224] Example: There are three mission segments under the hollow profile, and each mission segment has five flight types. Therefore, for the hollow profile, 5 mission profile spectra are needed, representing 5 flight types, respectively. Denoted as A1, B1, C1, D1, and E1. When compiling the A1 spectrum, arrange the spectra of flight type A under each mission segment, i.e., f 1,j,k ,f 2,j,k ,f 3,j,k , which can obtain the A1 spectrum. The same applies to other kinds of mission profile spectra.

[0225] (3) Compile the fly-continue-fly spectrum

[0226] Let y j,k be the number of times that the kth flight type(k=1,2...K) of the jth profile(j=1,2...L) appears in 1000 flights. Compile all yj,k In vector form, it is expressed as

[0227] A = [y 1,1 …y 1,K y 2,1 …y 2,K y 3,1 …y 3,K ……y L,1 …y L,K ]

[0228] In the formula, A is a vector of the number of occurrences of various task profiles. Among them The A vector and the B vector are in a one-to-one correspondence, indicating how many times the task profile appears in this 1000-time block profile. Then the total profile can be expressed as the task profile sequence G:

[0229] G = A * B = [F 1,1 F 1,1 F 1,1 F 1,1 ……F 1,2 F 1,2 F 1,2 F 1,2 ……F L,K F L,K F L,K F L,K ]

[0230] Among them, there are y 1,1 F 1,1 , y 1,2 F 1,2 , y L,K F L,K ;

[0231] Reordering the task profile under each complete flight in G can obtain the final flight-continuous-flight profile.

[0232] The specific method is as follows: first, number all the task profiles of G in order with a natural number sequence, and then correspond the number to a random integer sequence "x is " respectively, and sort according to the value of x is (For example, if the corresponding x is is 5, the corresponding task profile is placed in the 5th position) Thus, the final load spectrum is obtained.

[0233] (4) Random sorting method

[0234] The random sequence "x i " of the random alternating selection of peak, valley value and total load spectrum G in the above is realized by using the pseudo-random method of multiplication congruence method, and reasonable random results are given by adjusting the random parameters.

[0235] Multiplication congruence method:

[0236] y i+1 = λy i (mod2 k ) (1)

[0237] x i = y i / 2 m (2)

[0238] Firstly, the "y i " series is formed according to formula (1), and then the random series "x i " is obtained according to formula (2). The x i is calculated in the order of the total series arranged from small to large x is , so that the random integer series "x is " is obtained.

[0239] As to the parameter values in formula (1) and (2), the initial value y1 is an odd number; the coefficient λ = 8t-3, t is any natural number, k is determined according to the random period "n" as required, so that 2 k-2 ≥ n; m is 2. By changing the size of y1, t and k, different random series can be obtained.

[0240] The method for preparing the average gust spectrum will be further explained as follows:

[0241] 1. Determination of average cumulative overshoot number curve

[0242] For each level load of each mission segment, the overshoot number with a coverage rate of 50% is calculated, that is

[0243]

[0244] After the (Δn y , ΔN 50 ) i data pair is obtained, ΔN 50i is accumulated to obtain the average gust overload cumulative overshoot number curve corresponding to each mission segment.

[0245] 2. Preparation of average gust spectrum

[0246] According to the above method, the average gust spectrum is prepared.

[0247] The present application will be further explained according to examples as follows:

[0248] 1. Mission profile

[0249] A certain type of aircraft has one mission profile, which is mid-air flight. The preparation is in units of 1000 takeoffs and landings.

[0250] Table 2. Each task section of the empty profile

[0251] Task segment Time (min) Climb 2.1 Hole flight 10.1 Descent 2.3

[0252] 2. Measured data of load spectrum

[0253] The measured flight parameter data of 10 takeoffs and landings of X aircraft are measured, and the flight parameter data provided include time, flight altitude, y-direction overload at the center of gravity, left / right remaining fuel, flap deflection angle, elevator deflection angle.

[0254] 3. Preprocessing of measured load data

[0255] (1) n of overload data y Standardization processing

[0256] The standard aircraft mass data of each task section is shown in the table below, and the true mass data of the load is calculated by subtracting the fuel consumption from the mass of the aircraft. The fuel consumption is obtained by total average (the time for calculating the average fuel consumption starts from the beginning of takeoff taxiing to the end of landing impact).

[0257] The y-direction overload data of the center of gravity of the aircraft is standardized.

[0258] Table 3. Standard mass of each task section

[0259] Task segment Total mass (kg) Climb 22500 Hole flight 20000 Descent 15000

[0260] (2) Peak and valley value collection

[0261] Before counting and statistics, the peak and valley values of the statistical parameters are detected according to the method of section 4.2.3 to obtain the gust overload peak and valley value data pair (Δn y峰 , Δn y谷 ) i .

[0262] 4. Statistical analysis of overload cumulative exceedance number

[0263] (1) Overload time history counting

[0264] Taking an interval of 0.05g starting from Δn y =0.05g. The overload time history counting is performed on the standardized overload time history, and the gust overload peak and valley value data pair (Δn y峰 , Δn y谷 ) i is obtained, where the peak value is all positive overload and the valley value is all negative overload.

[0265] (2) Standardization processing of overload exceedance number considering time

[0266] The positive peak overload and negative valley overload are counted respectively. The flight duration is considered and the overload cumulative exceedance is normalized. The standard time of mission segment is shown in Table 2.

[0267] (3) Overload exceedance statistics of each mission segment

[0268] The distribution parameter of each level of overload exceedance is estimated, and the result is shown in Table 3. Figure 4-5

[0269] 5. Severe gust spectrum compilation

[0270] (1) Load spectrum damage distribution

[0271] The load spectrum damage distribution under the hollow mission profile is shown in Table 4. Figure 6

[0272] The Weibull distribution parameters of load spectrum damage under the hollow mission profile are shown in Table 5.

[0273] Table 5 Weibull distribution parameters

[0274] α β 0.75 1.30

[0275] (2) Severe spectrum severity determination

[0276] In this embodiment, the severe spectrum severity is taken as 90%

[0277] (3) Exceedance coverage P t determination

[0278] For the hollow mission profile, the corresponding u p = 1.196

[0279] (4) Severe cumulative exceedance curve determination

[0280] The severe cumulative exceedance curve of the climb mission segment under the hollow mission profile is shown in Table 6. Figure 7

[0281] (5) High load clipping and low load clipping

[0282] 1) High load clipping

[0283] The high load clipping value of each mission segment is shown in Table 7.

[0284] Table 7 High load clipping value of each mission segment

[0285] Task segment High load intercept / g Hole climb 0.86 Hole level flight 0.98 Hole descent 0.99

[0286] 2) Low load clipping

[0287] The low load clipping value of each mission segment is shown in Table 8. ​​​

[0288] Low load deletion values for each mission segment of Table 6

[0289] Task segment Low load delete / g Hole climb 0.16 Hole level flight 0.17 Hole descent 0.22

[0290] (6) Load spectrum compilation

[0291] 5x5 spectrum compilation

[0292] According to the compilation principle, the flight type number of the flight profile under the severe wind gust spectrum is given respectively, the flight type number of the severe wind gust spectrum is shown in Table 7, and the wind gust spectrum of the climbing mission segment under the hollow flight profile is shown in Table 8.

[0293] Table 7 Flight type number of typical flight profile (1000 flights)

[0294]

[0295] Table 8 Hollow flight climbing wind gust spectrum

[0296]

[0297] Random arrangement of fatigue load spectrum

[0298] According to the above principle, the fatigue load spectrum (load time sequence) is compiled, and the random spectrum of A flight type is shown as Figure 8 .

[0299] The present application simulates the wind gust load time history family of the mission profile of the aircraft group by the Monte Carlo method, and then calculates the damage of each mission profile and its corresponding distribution (according to the double-parameter Weibull distribution, the coverage rate is 90%), calculates the severe damage of each profile, and then according to the severe damage, the coverage rate corresponding to the number of exceedances of the specified overload of each mission segment under the mission segment is inverted, that is, the damage corresponding to the overload cumulative exceedance number curve of the coverage rate of each mission segment under the mission profile should be equal to the severe damage of the mission profile.

[0300] Each embodiment in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same and similar parts between each embodiment can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the related parts can be referred to the method part.

[0301] The foregoing description of the disclosed embodiments enables a person skilled in the art to make or use the application. Modifications of these embodiments will occur to persons of skill in the art, and that the appended claims are intended to cover all such modifications that do not depart from the true spirit and scope of the application. Therefore, the application is not limited to the embodiments shown but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a measured severe wind spectrum based on a Monte Carlo method, characterized in that, The method comprises the following steps: S1. defining the aircraft mission profile: according to the use requirements of the aircraft, the composition and proportion of the mission profile are given, and for each mission profile, the mission profile parameters are defined; S2. load data preparation: the measured load data obtained according to the mission profile is preprocessed to obtain the load data; S3. statistical processing of load data: the distribution parameters of the exceedance number under the specified overload are estimated according to the maximum likelihood estimation function; S4. Calculate the damage distribution of each mission profile, and take the reliability P of the fleet L The 90% damage is taken as the serious damage; and then the serious damage is inverted to obtain the serious overload cumulative exceeding number curve under each mission segment: S41. Based on the distribution parameters of the overshoot number under the specified overload, the Monte Carlo method is used to randomly extract the time sequence of the sudden burst overload under the task profile (n y,max , n y,min ) i ; wherein i is the level of the overload. S42. Calculate the time series (n y,max , n y,min ) i of the single-cycle equivalent damage, and accumulate the load spectrum damage D eq ; S43. According to the load spectrum damage D eq Fitting D eqi and the corresponding empirical frequency value f i The Weibull distribution parameters a and b are calculated; where D eqi is the load spectrum damage corresponding to the i-th sample data; S44. Calculate the load spectrum reliability P L set to 90%, calculate the exceedance coverage P t The specific method is: D eq90 = exp((ln(-ln(1-0.9))+αlnβ) / α) The inversion obtains the current severe damage D eq90 The corresponding u pt And the corresponding coverage of the transcendental number P t Where u pt Is the quantile of the standard normal distribution corresponding to the coverage of the transcendental number P t ; For each level of load of each task segment, the coverage is calculated as P according to the distribution function and distribution parameters in S3 t a transcendental number, i.e. where μ i and σ i are the mean and standard deviation of the logarithm of the corresponding number of overloads n y,i i. After obtaining (Δn y , ΔN Pt ) i Data pairs, the corresponding each task segment severe wind overload cumulative overrun curve is obtained by fitting; S5. determining the 5-level load by the load spectrum equivalent criterion, and the severe gust overload spectrum is compiled by the root 5x5 spectrum compilation method: for the typical mission segment in the mission profile, the severe mission segment spectrum is compiled; the mission profile severe spectrum is composed of the mission segments; and the fly-continuous-fly spectrum with 1000 takeoffs and landings as a cycle is compiled by randomly sorting the mission profile severe spectrum in proportion.

2. The method of claim 1, wherein the method is characterized by, The specific content of S1 includes: determining the mission profile, the proportion of the mission profile and the composition of the mission profile of the aircraft, and giving the mission profile parameters, wherein the mission profile parameters include: mission segment, height, speed, weight, flight distance and flight time of the mission segment; and the division of the measured data mission segment, the division principle is: a) the climb start point is taken as the criterion when the flap deflection angle changes to 0°; b) the climb end point / flat start is taken as the criterion when the turning point of the height curve turns flat; c) the flat end point / gliding start point is taken as the criterion when the turning point of the height curve descends; d) the glide end point is taken as the criterion when the flap deflection angle changes to 35°.

3. The method of claim 1, wherein the method is characterized by, The specific content of pre-processing the measured load data in S2 includes: (1) Overload data n y Standardization processing Δn y0 = Δn yi *Gi / G0 where Δn y0 is the corrected overload value according to the standard mission segment; Δn yi is the measured overload value, G0 is the standard aircraft mass for the mission segment according to the flight profile, and G is the actual aircraft mass. i is the measured overload value, G0 is the standard aircraft mass for the mission segment according to the flight profile, and G is the actual aircraft mass. (2) separation of gust and maneuver overloads The maneuver and gust loads contained in the measured load data are separated, and the interpretation conditions of the maneuver and gust load spectrum in the air flight are: a) the overload changes slowly and the duration is more than 2s, which is a maneuver, otherwise it is a gust; b) the overload with a rudder deflection angle absolute value equal to or greater than 4° is a maneuver overload, otherwise it is a gust overload; c) for the climb and glide mission segments, since there are fewer maneuver actions, the loads are approximately considered as gust loads, and the separation of gust and maneuver is not performed again; (3) peak and valley value collection In the counting process, all the peak and valley values are detected and represented in order as Δn zi , the data between the peak and valley values are filtered out while the corresponding sampling point serial numbers are kept, if the following conditions are met: or then take the peak or valley value once.

4. The method of claim 1, wherein the method is based on a Monte Carlo method. S3 also includes: (1) the overload time history count adopts the limited cross-peak count method; with the 1g load state of each mission segment as the reference, when the valley value between two peaks or the peak value between two valleys does not exceed the deviation of the reference line of the mission segment: for positive overloads, the deviation is the upper deviation, and only the maximum peak value is recorded; for negative overloads, the deviation is the lower deviation, and only the minimum valley value is recorded; wherein the absolute values of the upper deviation and the lower deviation are both 20% of the maximum peak value; (2) obtaining the cumulative exceedance number of each level load; The peak value obtained by the cross-peak counting method is defined as a positive peak value and a valley value as a negative valley value. Several load levels are selected, and the cumulative exceedance counts of positive and negative gust overloads are recorded for both positive and negative peak values, denoted as N(+Δn). y ) and N(-Δn y ); acquisition + Δn y cumulative excess number: (3) considering the flight duration, the overload cumulative exceedance number is standardized; The measured data corresponds to the time of flight t M , the standard time of the task segment is t S , the cumulative overrun number N of the i-th level load obtained by the measured time ib , then N ib The standardization rule is: where N i is the cumulative number of exceedances of the ith level of load for the standard time; the cumulative exceedance number of each level load is standardized to obtain the mission segment overload cumulative exceedance number data of the standard time.

5. The method of claim 1, wherein the method is based on a Monte Carlo method. The distribution parameter estimation method of the exceedance number under the specified overload in S3 is: the maximum likelihood estimation method is used to estimate the distribution parameters; the likelihood function is: wherein x i = lg ΔN i ; the target parameters (μ, σ) are solved by taking the derivative of the likelihood function with respect to μ and σ as zero 6. The method of claim 1, wherein the method is based on a Monte Carlo method. The specific content of S41 includes: 1) Randomly draw a random number u from a standard normal distribution p , u p denotes the quantile of the standard normal distribution for the corresponding coverage P, the i-th level of the stormer overload Δn yi the corresponding coverage P is the transfinite number ΔN P , i is: 2) on ΔN i Cumulative counts, resulting in cumulative excesses Thus, from (Δn y ,N) i Data pairs are fitted to obtain Δn y Excess curve; 3) Repeat step 1) for all mission segments to obtain the gust overload exceedance curves for all mission segments of the mission profile represented by this sample; 4) Discretize the measured burst overload cumulative overrun number curve family to obtain the overrun number data pairs (±Δn y* ,ΔN) ij corresponding to each level of burst overload, where i is the level of overload, and j is each task segment; wherein the discretization process is: calculate to obtain the overrun number ΔN i =N i -N i-1 , obtain the overrun number data pairs (Δn y ,ΔN) i corresponding to different burst overloads, i=1,…,n; 5) summing the specified Δn of each task segment under the task segment of the burst overload, respectively, to obtain total overrun data pair of each level of burst overload y m is the number of task segments;​ 6) For the burst overload, the positive and negative overloads of each level are sorted by the absolute value and matched, and the time sequence of the burst overload of the task profile under this sampling is obtained (n y,max , n y,min ) i ; 7) Repeat the sampling process for each mission profile to obtain a family of gust overload versus time sequences for each mission profile.

7. The method of claim 6, wherein the method is based on a Monte Carlo method. 2) Δn for single landing fit y The specific method for the transfinite curve is: The overload accumulation excess number data pair (Δn y ,N) i is fitted, and the fitting equation is: Δn y = a * lgN + b The cumulative exceedance curves of each mission segment are calculated.

8. The method of claim 1, wherein the method is based on a Monte Carlo method. The specific content of S42 includes: According to the prepared time series of the sudden overload n y,max , n y,min ) i , the single-cycle equivalent damage is calculated, and the load spectrum damage D eq is accumulated, and the specific method is as follows: where m is the damage exponent, i.e. the slope of the material / structure fatigue S-N curve at stress ratio R = 0, d eq is the damage per cycle.

9. The method of claim 1, wherein the method is based on a Monte Carlo method. S4 also includes: S45. High load cutting and low load cutting of the severe cumulative exceedance curve of the mission segment; (1) High load cutting Determine the level and quantity of high load in the severe gust spectrum, and select the high load of the preset frequency as the cutting value; (2) Low load cutting Delete the small amplitude load in the severe gust spectrum or convert it to a certain level of load with equal damage.

Images