5x5 spectral iterative method based on overload cumulative overshoot curve

Abstract

The application discloses a 5*5 spectrum iteration compiling method based on overload cumulative overrun number curve, which comprises the following steps: step 1, determining an aircraft mission profile; step 2, collecting an overload overrun number curve and pre-processing; step 3, collecting and counting peak and valley values of the pre-processed overload overrun number curve to obtain an overload cumulative overrun number curve; step 4, according to the overload cumulative overrun number curve and 5*5 spectrum compiling principle, a set of 5*5 spectrum iteration compiling method is constructed, and 5*5 spectrums of each mission section of the mission profile are compiled; step 5, compiling a mission section spectrum according to the 5*5 spectrums of each mission section; step 6, compiling a mission profile spectrum according to the mission section spectrum and the aircraft mission profile; and step 7, compiling a fly-continue-fly spectrum. The method can not only better meet the 5*5 spectrum compiling principle and reduce the iteration time of 5*5 spectrum compiling, but also reflect the damage characteristics and equivalence of the load and take into account the convenience and speed characteristics.

Description

Technical Field

[0001] This invention relates to the field of flight spectrum compilation technology, and more specifically to a 5×5 spectrum iteration compilation method based on overload cumulative exceedance number curves. Background Technology

[0002] Currently, the load spectrum refers to the spectrum compiled from the load time history experienced by the aircraft during flight and ground missions. The measured flight load spectrum of the aircraft is compiled through specialized testing modifications and flight tests. It is used to determine and verify the design service life of the aircraft and is a prerequisite for determining the fatigue life extension of the aircraft structure.

[0003] Different design criteria require the development of different types of load spectra. Aircraft structural design has gone through several development stages, including static strength design, safe life design, damage tolerance design, and durability design. Static strength design does not require the development of load spectra, and early safe life design (before the 1960s) could basically meet the design requirements by developing block load spectra. After the 1970s, safe life design and durability design were often considered together with damage tolerance design. At this time, it was necessary to develop flight-continue-flight load spectra for testing and analysis.

[0004] Regarding specific flight-continued-flight spectrum compilation methods, the most widely accepted civil aircraft spectrum compilation technique is the TWIST spectrum compilation method based on overload exceedance number curves. TWIST stands for TransportWinding Standard. It was first proposed by aviation scientists in the Netherlands and Germany in the late 1960s and early 1970s to solve the standardization problem of fatigue test spectra for transport aircraft wings, and has been successfully applied in full-scale fatigue tests of some Airbus and Boeing aircraft models. The TWIST spectrum generally needs to meet two principles:

[0005] (1) The principle of log-normal distribution of extreme values

[0006] (2) Load similarity principle for various flight types

[0007] Subsequently, Boeing proposed a 5×5 spectrum based on mission segments, which follows the same principles as the TWSIT spectrum. This method has since been adopted for the spectrum compilation of Boeing's commercial aircraft models.

[0008] Currently, the fitting and iteration methods for 5×5 spectrum / TWIST spectrum are too cumbersome. Taking Pan Qingrong's method as an example, it is necessary to use trial and error to determine the initial value, determine the range of the root variance σ and the mean a, and then perform trial and error again to determine the occurrence frequency of each flight type in order to satisfy the principle of extreme value log-normal distribution. The whole process is quite cumbersome and requires a lot of iterations.

[0009] Therefore, how to reduce the complexity of load spectrum compilation is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0010] In view of this, the present invention provides a 5×5 spectrum iterative compilation method based on the overload cumulative exceedance number curve. On the basis of the original 5×5 spectrum compilation method, the load equivalent damage principle is introduced, and a more convenient and faster 5×5 spectrum compilation method based on the overload exceedance number curve with fewer iterations is proposed. It can not only better meet the 5×5 spectrum compilation principle and reduce the 5×5 spectrum compilation iteration time, but also reflect the damage characteristics and equivalence of the load, while taking into account the convenience and speed.

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] A 5×5 spectrum iteration method based on overload cumulative exceedance number curves includes the following steps:

[0013] Step 1: Generate an aircraft mission profile based on the usage requirements of the aircraft under test, and determine the sequence, time ratio, parameters, and total number and ratio of aircraft mission profiles under a block spectrum.

[0014] Step 2: Collect the measured load data of the aircraft under test to generate the overload exceedance curve, and perform preprocessing, including separation and filtering of gust load and maneuver load;

[0015] Step 3: Collect and count the peak and valley values ​​of the preprocessed overload exceedance number curve to obtain the cumulative overload exceedance number curve, calculate the load level ratio of the cumulative overload exceedance number curve for each task segment, and segment the curve according to the load level ratio.

[0016] Step 4: Calculate the equivalent loads of each level of the cumulative overload exceedance curve for each mission segment and fit them to obtain the correlation coefficients. If the correlation coefficients are less than the set coefficient threshold, return to Step 3 to adjust the load level ratio; otherwise, obtain the 5×5 spectrum of each mission segment that constitutes the aircraft mission profile with the equivalent loads of each level.

[0017] Step 5: Compile the task segment spectrum based on the 5×5 spectrum of each task segment;

[0018] Step 6: Compile a mission profile spectrum based on the mission segment spectrum and the aircraft mission profile;

[0019] Step 7: Sort the mission profile spectrum and compile the flight-continued-flight spectrum, which is a 5×5 spectrum.

[0020] Preferably, step 1, determining the mission profile, includes determining the composition and proportion of the mission profile according to the aircraft's usage requirements, determining the composition, sequence, proportion, and parameters of the mission segments in each mission profile, and determining the total number and proportion of mission profiles under a block spectrum; the parameters of the mission profile include mission segments, mission segment altitude, speed, weight, flight distance, flight time, etc.

[0021] The preferred principle for dividing task segments is as follows:

[0022] The starting point of the climb is determined by the flap deflection angle becoming 0°.

[0023] The point at which the climb ends (the point at which level flight begins) turns into a flat-out state is used as the criterion.

[0024] The end point of level flight is determined by the inflection point of the descent curve;

[0025] The descent end point is determined by the flap deflection angle becoming 35°.

[0026] Preferably, the measured load data includes gust load and maneuver load, generating overload exceedance curves. Preprocessing includes separating gust load and maneuver load and standard time accumulation processing. Step 2, in which preprocessing, obtains the task segment overload accumulation exceedance data for each load level over the standard time, as well as the maneuver load data and gust load data. The process includes:

[0027] Step 21: Construct the standard time cumulative overload overload data for the task segment from the standard time cumulative overload overload numbers of each load level in the overload overload overload number curve;

[0028] Step 22: The separation conditions for separating the measured load data, including the motor load data and the sudden wind load data, are as follows:

[0029] Overload changes that are slow and last for more than 2 seconds are considered dynamic load data; otherwise, they are considered sudden wind load data.

[0030] An overload with an absolute value of 4° or greater on the control surface deflection is considered a maneuver overload; otherwise, it is considered a maneuver overload.

[0031] Preferably, the specific process of collecting and counting peak and valley values ​​of the preprocessed overload exceedance data curves for each mission segment of the aircraft mission profile to obtain the cumulative overload exceedance number curve for each mission segment, and segmenting the curves according to the load level ratio, is as follows:

[0032] Step 31: Collect peak and valley values ​​for the separated gust load data and maneuver load data for each task segment to obtain the corresponding gust overload peak and valley value data and maneuver overload peak and valley value data for that task segment;

[0033] Step 32: For both the separated peak-to-valley data of sudden wind overload and the peak-to-valley data of motor overload, time history counting is performed using the restricted cross-peak counting method to obtain the peak and valley values; the requirements for time history counting are as follows:

[0034] For positive overload, the 0g load state of each task segment is used as the baseline. When the valley value between two peaks or the peak value between two valleys does not exceed the upper deviation of the baseline of that task segment, the maximum peak value is recorded.

[0035] For negative overload, the 0g load state of each task segment is used as the baseline. When the valley value between two peaks or the peak value between two valleys does not exceed the lower deviation of the baseline of the task segment, the maximum valley value is recorded.

[0036] The upper and lower deviations are 20% of the maximum peak value;

[0037] Step 33: Select several load levels, and for the gust load data and maneuver load data, calculate the cumulative exceedance number N(+Δn) of the positive flight overload corresponding to the peak value. y The negative flight overload cumulative exceedance number N(-Δn) corresponding to the valley value. y The average of the cumulative exceedance numbers corresponding to the peak and valley overload values ​​is calculated as the overload Δn. y The cumulative transcendence number N(Δn) y ), represented as:

[0038]

[0039] Obtain the overload cumulative exceedance data pair (Δn) y N(Δn) y )) i , i represents the i-th level of overload;

[0040] Step 34: Standardize the cumulative exceedance number of the specified load, that is, convert the measured cumulative exceedance number data of the task segment overload into the cumulative exceedance number data of the standard time, and obtain the standard cumulative exceedance number N of the i-th level load at the standard time. i , is represented as:

[0041]

[0042] Among them, t M t represents the flight time corresponding to the measured payload data. S N(Δn) represents the standard time for the task segment. yi ) represents the cumulative exceedance count of the i-th level load obtained from the measured time; step 34 is performed on the cumulative exceedance count of all loads to obtain the standard overload cumulative exceedance count data pair (Δn) of the i-th level load for all takeoffs and landings of the i-th level load in each task segment. y ,N) i ;

[0043] Step 35: For each mission segment of the i-th load, generate a standard cumulative overload exceedance data pair (Δn) for all takeoffs and landings. y ,N) i The cumulative overload exceedance number N for this mission segment under a single takeoff and landing is obtained by summing and averaging. t , represented as:

[0044]

[0045] Where, N i,j N represents the standard cumulative exceedance number of the mission segment under the i-th level load during the j-th takeoff and landing; ti This represents the average cumulative overload exceedance number for this mission segment under a single takeoff and landing of the i-th level load; j represents the takeoff and landing sequence number; m represents the total number of actual takeoffs and landings.

[0046] Step 36: Based on the cumulative overload exceedance number of a single takeoff and landing for this mission segment under the above formula, obtain the cumulative overload exceedance number data pair (Δn) for a single takeoff and landing under the i-th level load. y N t ) i The cumulative overload exceedance number (Δn) for a single takeoff and landing under the i-th level load. y N t ) i By performing fitting, the cumulative overload exceedance number curve for each task segment is obtained, expressed as:

[0047] Δn y =a*lgN t +b

[0048] a and b both represent the parameters of the linear fit;

[0049] Step 37: Discretize the overload cumulative exceedance curve of each task segment into 4 segments, and represent them with x-axis pairs according to the low load cutoff value and the high load cutoff value, respectively represented as (LB5,LB4), (LB4,LB3), (LB3,LB2) and (LB2,LB1), where LB5 is the low load cutoff value, LB1 is the high load cutoff value, and LB2, LB3, and LB4 are the set values ​​between the low load cutoff value and the high load cutoff value, respectively;

[0050] Step 38: Calculate the load level ratio R for each discrete curve segment. i The expression is:

[0051] R i =(LB) i -LB i+1 ) / (LB1-LB5)

[0052] i = 1, 2, 3, 4;

[0053] Step 39 uses m straight lines to replace each discrete curve segment. The linear equation of the i-th discrete curve segment is expressed as:

[0054] Δn y =a i lgN t +b i

[0055]

[0056]

[0057] Among them, a i b i All are constants of the i-th segment of the load spectrum curve; Δn yd Equivalent load; N eq The equivalent load cycle number is denoted by S; S is the slope parameter of the material's SN curve; first, the discrete curve of each segment is fitted according to the linear equation to obtain the value of a for each segment. i b i Then calculate A as expressed by the formula. i Then A i Substitute into N eq In the calculation formula;

[0058] Step 310: Obtain the equivalent load Δn of the four discrete curves corresponding to the four levels of wind load data and motor load data, respectively. ydi (i = 2, 3, 4, 5) and the corresponding number of load cycles N eqi (i = 2, 3, 4, 5); the highest load level LB5 (Δn) yd1 As a high load constituting a level 5 load, its corresponding load cycle number 1 is taken as the corresponding occurrence number. High load refers to the high load intercepted by the high load, which is the highest level load in the 5×5 spectrum.

[0059] Preferably, the specific process for compiling the 5×5 spectrum for each task segment of the gust load data and the maneuver load data in step 4 is as follows:

[0060] Step 41: Determine the frequency y of each flight type i ;

[0061] Step 42: Determine the initial equivalent load for each stage and the number of cycles for each initial equivalent load;

[0062] Set an initial load level ratio to determine the starting point of the four discrete curves;

[0063] Take the initial load Δn at each stage ydoi Calculate the initial equivalent load cycle number N for each stage, using the average value of each discrete curve interval. eqoi ;

[0064] Step 43: Determine the frequency of each load level under each flight type;

[0065] Based on the initial equivalent load at each stage and the number of cycles of the initial equivalent load at each stage, and in accordance with the load similarity principle for each flight type, the number of occurrences of each load stage B under each flight type is determined. ij And the corresponding cumulative number of incremental overload cycles N at each level eqci ;

[0066] Ensure the cumulative number of cycles N for incremental overload at each level eqci Number of cycles N of the initial equivalent load eqoi equal;

[0067] Step 44: Analyze the occurrence frequency (y) of each flight type i The cumulative number of flights (BE) for each flight type is obtained by summing them up. i ,

[0068]

[0069] Step 45: Based on cumulative flight count (BE) i The geometric mean method is used to obtain the transcendental number degree (LDE). i ;

[0070] LDE i = (BE) i *BE i-1 )^0.5;

[0071] Step 46: Divide the number of transcendences by the total number of rises and falls N corresponding to the block spectrum. c The exceedance probability P corresponding to each load level is obtained. i

[0072] P i =LDE1 / N c ;

[0073] Step 47: Calculate the equivalent load Δn for each stage ydi The probability of exceeding P i and the corresponding standard normal distribution quantile u p,i The data pair (lgΔn) is obtained. ydi ,u p,i (i = 2, 3, 4, 5); data pairs (lgΔn) ydi ,u p,i The correlation coefficient r is obtained by fitting the following formula;

[0074] lgΔn ydi =μ+u p,i σ

[0075] μ represents the mean, and σ represents the standard deviation;

[0076] Step 48: Determine if the correlation coefficient is greater than the set coefficient threshold. If it is, output the 5×5 spectrum for each task segment; otherwise, return to step 36. Adjust the load level ratio R. i If the correlation coefficient is greater than 0.99, then the 5×5 spectrum satisfies the principle of log-normal extreme value distribution, thus yielding the final Δn. ydi .

[0077] Preferably, the specific process of step 5 is as follows:

[0078] Step 51: Based on the 5×5 spectrum of each mission segment, link the payloads at each level and their corresponding frequencies, randomly and alternately select peak and valley values ​​for random pairing, and compile payload pair sequences f for all flight types under all mission segments. i,j,k Obtain the mission segment spectrum F of the k-th flight type from the j-th mission profile. j,k ;

[0079] Step 52: Based on the mission segment spectrum, sort the mission segments i in the mission profile according to the order i = 1, 2, 3... p to obtain a complete flight mission profile spectrum f. 1,j,k ,f 2,j,k ,f 3,j,k ,f 4,j,k ...f p,j,k ;

[0080] Step 53: Obtain the task profile spectrum of all types by sorting, and represent it in vector form as follows:

[0081] 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 ], L represents the number of task segments under the task profile, and K represents the number of task profiles.

[0082] Preferably, the specific process of step 6 is as follows:

[0083] Step 61: Categorize all types of mission profiles according to flight type;

[0084] Step 62: Randomly arrange the flight-continued-flight patterns according to the frequency of occurrence of each flight type under the block spectrum to form the flight-continued-flight pattern.

[0085] Preferably, the mission profile spectrum consists of several mission segment spectra; the mission profile spectra are randomly sorted proportionally to compile a flight-continue-flight spectrum with a block spectrum as the period.

[0086] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a 5×5 spectrum iterative compilation method based on the overload cumulative exceedance number curve. On the basis of the original 5×5 spectrum compilation method, the load equivalent damage principle is introduced, providing a more convenient and faster 5×5 spectrum compilation method based on the overload exceedance number curve with fewer iterations. It can not only better meet the 5×5 spectrum compilation principle, but also reflect the damage characteristics and equivalence of the load, while also taking into account convenience and speed. Attached Figure Description

[0087] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0088] Figure 1 The attached figure is a schematic diagram of the 5×5 spectrum iteration compilation method based on the overload cumulative exceedance number curve provided by the present invention;

[0089] Figure 2 The attached figure is a schematic diagram of peak-valley value detection in an embodiment provided by the present invention;

[0090] Figure 3 The attached figure is a schematic diagram illustrating the limitation of peak count across averages in an embodiment provided by the present invention;

[0091] Figure 4 The attached figure is a schematic diagram of the overload exceedance number curve discretization in the embodiment provided by the present invention;

[0092] Figure 5 The attached figure is a schematic diagram of the cumulative overload exceedance curve of the climbing task section in section 1 of the embodiment provided by the present invention;

[0093] Figure 6 The attached figure is a schematic diagram of the extreme load distribution in the 5×5 spectrum in the embodiment provided by the present invention;

[0094] Figure 7 The attached figure is a schematic diagram of the random spectrum of the climb mission segment of flight type A in section 1 of the embodiment provided by the present invention. Detailed Implementation

[0095] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0096] This invention discloses a method for iterative compilation of a 5×5 spectrum based on an overload cumulative transcendence number curve, comprising the following steps:

[0097] S1: Determine the mission profile; according to the aircraft's usage requirements, provide the composition and proportion of the mission profile; for each mission profile, clarify the composition, order, proportion, and parameters of the mission segments that constitute the mission profile; determine the total number and proportion of mission profiles under a block spectrum; for the compilation of measured spectrum, the typical mission profile of the aircraft should be obtained by simplifying and classifying the types and proportions of missions performed by the aircraft in actual use; for the compilation of design spectrum, the typical mission profile of the aircraft should be determined based on the aircraft design and expected usage.

[0098] Determine the aircraft's mission profile, mission profile scale, and mission profile composition, and provide mission profile parameters such as mission segments, mission segment altitude, speed, weight, flight distance, and flight time. Divide the measured data mission segments according to the following principles:

[0099] a) The starting point of the climb is determined by the flap deflection angle becoming 0°;

[0100] b) The point at which the climb ends (the point at which level flight begins) turns flat is used as the criterion.

[0101] c) The end point of level flight is determined by the inflection point of the altitude descent curve;

[0102] d) The descent end point is determined by the flap deflection angle becoming 35°;

[0103] The overload spectrum for the mission segment is compiled according to the “5×5” spectrum, that is: first, the overload spectrum is discretized into five levels of spectrum, and then the discrete spectrum is compiled into load spectra of five typical flight types with different degrees of lightness and heaviness; among them, all loads in the climb and glide mission segments are uniformly treated as gust loads.

[0104] S2: Obtain the overload cumulative exceedance number curve;

[0105] The overload exceedance number curve can be referenced from the cumulative exceedance number curve of gust speed collected by NACA in the past or the overload exceedance number curve recorded by the flight parameter recorder of similar aircraft in the past. Alternatively, it can be obtained by statistical processing of the measured overload time series of this type of aircraft. Specifically, it should include the following steps:

[0106] S21: Separation of maneuverability and gust overload;

[0107] The measured load data during the flight mission segment includes maneuver and gust loads, which need to be separated. The interpretation of the in-flight maneuver and gust load spectra is determined by comprehensively considering the following three conditions:

[0108] a) If the overload changes slowly and lasts for more than 2 seconds, it is considered a maneuver; otherwise, it is considered a gust.

[0109] b) An overload with an absolute value of 4° or greater on the control surface deflection angle is a maneuver overload; otherwise, it is a maneuver overload.

[0110] The following explanation uses the compilation of a sudden wind overload spectrum as an example. The method for compiling a motor overload spectrum is the same as that for compiling a sudden wind overload spectrum.

[0111] S22: Peak-valley value collection;

[0112] Peak-valley detection involves identifying all peak and trough values ​​during the counting process, filtering out the data between the peaks and troughs, and retaining the corresponding sampling point numbers; for example... Figure 2 As shown, retain Δn z1 , Δn z3 , Δn z5 , Δn z6 , Δn z7 , Δn z8 , Δn z10 Peak and valley points, remove Δn z2 , Δn z4 , Δn z9 Sampling points;

[0113] If satisfied

[0114]

[0115] Then the peak (or valley) value is taken once;

[0116] S23: Overload time history count;

[0117] The time history of sudden wind overload after peak and valley value acquisition is counted using the restricted cross-average peak counting method. Specific processing requirements are as follows: Using the "0g" load state of each task segment as the baseline, if the valley value (or peak value between two valleys) does not exceed the upper deviation (lower deviation for negative overload) of the "0g" baseline for that task segment, only the maximum peak value (valley value for negative overload) is recorded. The upper and lower deviations are 20% of the maximum peak value. This deviation is the constraint condition, and the cross-average peak counting is as follows: Figure 3 As shown;

[0118] S24: Obtaining the cumulative exceedance number of each load level;

[0119] The peak values ​​obtained by the above-mentioned peak-to-average count method are greater than 0, and the valley values ​​are less than 0. Several load levels are selected, and the cumulative exceedance counts of positive flight overload (gust overload and maneuver overload) and negative flight overload are recorded for both positive peak values ​​and negative valley values, denoted as N(+Δn). y ) and N(-Δn y );

[0120] In engineering, the geometric mean is used to obtain the average of the cumulative exceedance numbers corresponding to a symmetrical sudden wind overload, thus obtaining +Δn. y The family of cumulative transcendental number curves is shown in the following equation:

[0121]

[0122] Construct overload cumulative exceedance data pairs (Δn) y N(Δn) y )) i , i represents the i-th level of overload; for ground landing impact overload, the cumulative exceedance number of positive overload and the cumulative exceedance number of negative overload are processed separately;

[0123] S25: Overload data n y Standardized processing;

[0124] Since the measured load spectrum data differs between the mission segment flight time and the standard time of the mission segment in the mission profile, it is necessary to convert the measured cumulative overload exceedance data of the mission segment into cumulative exceedance data of the standard time. Let the flight time corresponding to the measured data be t. M The standard time for the task segment is t. S The cumulative exceedance number N(Δn) of the i-th level load obtained from the measured time. yi If the cumulative exceedance number of the i-th level load at standard time is:

[0125]

[0126] By substituting the cumulative overload exceedance counts for each load level according to the above formula, the cumulative overload exceedance count data for the task segment at standard time can be obtained;

[0127] S26: Fitting of the overload cumulative exceedance number curve;

[0128] For each mission segment, the cumulative overload exceedance data pair (Δn) for all takeoffs and landings. y ,N) i (Sequence number i, overload level sequence number) are summed and averaged to obtain the cumulative overload exceedance data pair (Δn) for this mission segment under a single takeoff and landing. y N t ) i (Sequence number j represents the takeoff and landing sequence number), where

[0129]

[0130] Where, N i,j N represents the cumulative exceedance count of the i-th level overload of the task segment during the j-th takeoff and landing; ti The number of overloads exceeded in a single takeoff and landing under the i-th level of overload is the average number of overloads in this mission segment; j represents the takeoff and landing sequence number; m represents the total number of takeoffs and landings measured.

[0131] For data pairs (Δn) y N t ) i The fitting equation is as follows:

[0132] Δn y =a*lgN t +b

[0133] a and b both represent the parameters of the linear fitting; the cumulative overload exceedance curves for each task segment are obtained, including the cumulative overload exceedance curves for gust overload and the cumulative overload exceedance curves for maneuver overload (positive / negative);

[0134] S3: 5×5 spectrum compilation;

[0135] The 5×5 spectrum based on the mission segment is shown in Table 1. The vertical axis of the 5×5 spectrum represents the five flight types, and the horizontal axis represents the five payload levels. The number of payload levels included in each flight type is B. ij The number of occurrences of each flight type is y. i The specific compilation process is as follows:

[0136] Table 1 5×5 spectrum

[0137]

[0138]

[0139] S31: High load cut-off and low load cut-off;

[0140] The cumulative exceedance curve of the task segment determined by S2 is subjected to high-load truncation and low-load truncation.

[0141] (1) High load interception

[0142] The high load that occurs once in 1000 flights in this mission segment is taken as the cutoff value;

[0143] (2) Low-load cutoff

[0144] The spectrum usually contains a large number of small load cycles, which need to be deleted or converted into a certain level of load for equivalent damage; usually, the fatigue limit of the key parts is determined based on the above analysis, and the overload value corresponding to 70% to 80% of the fatigue limit is removed.

[0145] S32: Equivalent load method;

[0146] Based on the cumulative exceedance number curve of the task segment overload, the load spectrum equivalent is calculated;

[0147] Calculation of equivalent load: Discretize the cumulative overload exceedance curve into 4 segments. These four straight lines are represented by x-axis pairs, namely (LB5, LB4), (LB4, LB3), (LB3, LB2), and (LB2, LB1), as follows. Figure 4 As shown;

[0148] Where LB5 is the low-load cutoff value, LB1 is the high-load cutoff value, and the load level ratio R is defined. i (i = 1, 2, 3, 4).

[0149] R i =(LB) i -LB i+1 ) / (LB1-LB5)

[0150] Suppose that a discrete segment for which equivalent calculation is required is represented by m straight lines, and its equivalent load is Δn. yd The equivalent load cycle number is N eq The linear equation for the i-th segment is:

[0151] Δn y =a i lgN t +b i

[0152] In the formula, a i b i Let be a constant of the i-th segment of the load spectrum curve;

[0153] Equivalent load cycle number

[0154]

[0155] In the formula:

[0156]

[0157] S is the slope parameter of the SN curve for the material; for aluminum alloy, S = 2.0.

[0158] Four discrete curves yielded four levels of equivalent load Δn. ydi (i = 2, 3, 4, 5) and the corresponding number of loops Neqi (i = 2, 3, 4, 5), high load LB5 (Δn) yd1 Together with the corresponding cycle number 1, they constitute the five levels of load and their corresponding occurrence counts;

[0159] S33: Compilation of 5×5 spectra;

[0160] The compilation of the 5×5 spectrum involves the following iterative process:

[0161] S331: Determine the frequency of each flight type; the frequency y of each flight type under this mission profile should be determined first. i The number of flight types in the 5×5 spectrum across different mission segments under the same mission profile should remain consistent. For the 5×5 spectrum, the setting of the number of flight types for each flight type in existing technologies with a cycle of 5000 flights can be referenced, as shown in the table below.

[0162] Table 2 Determination of the Number of Flights for Each Flight Type

[0163] Flight type frequency A 1 B 13 C 215 D 1067 E 3704

[0164] S332: Determine the initial load levels and number of cycles; First, set an initial load level ratio and determine the starting points of the four discrete segments. For the load level ratio of gust loads, refer to the table below.

[0165] Table 3 Reference values ​​for initial load level ratio

[0166] <![CDATA[Load level ratio R i > Gust of Wind Motor spectrum <![CDATA[R1]]> 0.425 0.35 <![CDATA[R2]]> 0.3 0.3 <![CDATA[R3]]> 0.15 0.2 <![CDATA[R4]]> 0.125 0.15

[0167] And take the initial load Δn for each stage ydoi This is the average value over the interval. Based on the load value Δn ydi and the probability of exceeding P i The calculation formula determines the number of cycles N of the initial equivalent load. eqoi Based on the load level ratio, four intervals are determined. Each interval has a maximum value and a minimum value. Their average value is the initial load at each level. Thus, the four intervals correspond to four initial loads.

[0168] S333: Determine B ij The number of times each load level occurs under each flight type should be determined based on the equivalent load and the number of equivalent load cycles, following the principle of load similarity for each flight type. Specifically, the cumulative overload exceedance curve fitted to the occurrence frequency of the five load levels under each flight type should maintain a similar shape. ij ), and the corresponding cumulative number of incremental overload cycles N at each level. eqci It should also be noted that, in principle, the cumulative number of iterations N for each incremental overload level should be guaranteed in this iteration. eqci Number of cycles N of the initial equivalent load eqoi They are equal, but differences of a single digit are allowed;

[0169] S334: Determine the exceedance probability P corresponding to the extreme loads of each flight type. i ;

[0170] First, the number of flights for each flight type (y) is used as an example. i The cumulative number of flights (BE) for each flight type is obtained by summing them up. i ,

[0171]

[0172] Based on cumulative flight count BE i The geometric mean method is used to obtain the transcendental number degree (LDE). i ;

[0173] LDE i = (BE) i *BE i-1 )^0.5

[0174] Divide the number of overshoots by the total number of rises and falls corresponding to the block spectrum N. c That is, to obtain the exceedance probability P corresponding to each load level. i

[0175] P i =LDE1 / N c

[0176] The extreme loads at each level and their corresponding exceedance probabilities are shown in the table below;

[0177] Table 4. Schematic diagram of extreme loads at each level and corresponding exceedance number probabilities.

[0178] Flight type Extreme load Number of flights Cumulative number of flights Exceeding the limit Exceeding probability A <![CDATA[Δn yd1 ]]> <![CDATA[y1]]> <![CDATA[BE1]]> <![CDATA[LDE1]]> <![CDATA[P1]]> B <![CDATA[Δn yd2 ]]> <![CDATA[y2]]> <![CDATA[BE2]]> <![CDATA[LDE2]]> <![CDATA[P2]]> C <![CDATA[Δn yd3 ]]> <![CDATA[y3]]> <![CDATA[BE3]]> <![CDATA[LDE3]]> <![CDATA[P3]]> D <![CDATA[Δn yd4 ]]> <![CDATA[y4]]> <![CDATA[BE4]]> <![CDATA[LDE4]]> <![CDATA[P4]]> E <![CDATA[Δn yd5 ]]> <![CDATA[y5]]> <![CDATA[BE5]]> <![CDATA[LDE5]]> <![CDATA[P5]]>

[0179] This leads to the acquisition of overload Δn at each level. ydi The probability of exceeding P i and the corresponding standard normal distribution quantile u p,i The data pair (lgΔn) is obtained. ydi ,u p,i (i = 2, 3, 4, 5);

[0180] S335: Determine Δn ydi (i = 2, 3, 4, 5);

[0181] From the transcendental probability P i The expression and N in S333 eqci Reverse calculation of overload Δn at each stage ydi Combining the exceedance probabilities of each overload level in S334 with the corresponding quantiles u p,i The data pair (lgΔn) is obtained. ydi ,u p,iFit the following formula to obtain the correlation coefficient r, and repeatedly modify the stress level ratio R. i (i = 2, 3, 4, 5), until the correlation coefficient is greater than 0.99, that is, the 5×5 spectrum is considered to satisfy the principle of extreme value log-normal distribution, thus obtaining the final Δn. ydi ;

[0182] lgΔn ydi =μ+μ p,i σ and μ represent the mean, and σ represents the standard deviation;

[0183] S336: Perform the above steps for all task segments and compile a 5×5 spectrum for all task segments;

[0184] S34: Random compilation of load spectrum;

[0185] S341: Compile the task segment spectrum;

[0186] Based on the data in the 5×5 spectrum compiled in S33, the payloads at each level and their corresponding frequencies are linked together. Peak and valley values ​​are randomly selected alternately (i.e., peak and valley values ​​are randomly selected) and randomly paired. In this way, the spectrum (payload pair sequence) of various flight types under all mission segments is compiled.

[0187] S342: Compile the mission profile spectrum;

[0188] Let F be the load spectrum of the k-th flight type on the j-th profile. j,k F j,k This can be represented as a load pair sequence f i,j,k Sort according to the task segment order i (i = 1, 2, 3... m) under this profile, i.e., f 1,j,k ,f 2,j,k ,f 3,j,k ,f 4,j,k ...f m,j,k This yields the mission profile spectrum for a complete flight. Using the method described above, mission profile spectra for all flight types under all mission profiles are compiled; all types of mission profile spectra are represented in vector form as follows:

[0189] 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 ]

[0190] If there are five mission segments under profile 1, and each mission segment has five flight types, then for the hollow profile, a total of 5 mission profile spectra need to be compiled, representing the 5 flight types, denoted as A1, B1, C1, D1, and E1. When compiling the A1 spectra, the spectra of flight type A under each mission segment are arranged together, i.e., f 1,j,k ,f 2,j,k ,f 3,j,k The A1 spectrum can be obtained by doing so, and the same applies to other types of task profile spectra.

[0191] S343: Compile flight-continued flight spectrum;

[0192] The flight patterns are randomly arranged according to the frequency of occurrence of each flight type under the block spectrum, forming a flight-continued-flight pattern.

[0193] In one specific embodiment, the process is as follows:

[0194] S1: Mission Profile;

[0195] Take a typical mission profile of a certain type of aircraft as profile 1, and compile a gust spectrum with 1000 takeoffs and landings as the unit, that is, the unit of the block spectrum is 1000 takeoffs and landings.

[0196] Table 5 Section 1 for each task segment

[0197] Task segment Standard weight (kg) Climb 22500 Flying 20000 decline 14500

[0198] S2: Measured load spectrum data;

[0199] A total of 50 takeoffs and landings of a certain aircraft were measured. The flight data provided includes time, flight altitude, Y-axis overload at the center of gravity, left / right engine remaining fuel, flap deflection angle, and elevator deflection angle.

[0200] S3: Determine the overload cumulative exceedance number curve;

[0201] The overload exceedance number curve can be referenced from the cumulative exceedance number curve of gust speed collected by NACA in the past or the overload exceedance number curve recorded by the flight parameter recorder of similar aircraft in the past, or it can be obtained by statistical processing based on the measured overload time series of this type of aircraft.

[0202] Determine the cumulative overload exceedance curves for all mission profiles. The cumulative overload exceedance curve for the gust overload during the climb mission in profile 1 is as follows: Figure 5 As shown;

[0203] S4: Perform high-load and low-load truncation on the overload cumulative exceedance curve;

[0204] (1) High load interception

[0205] The high-load intercept values ​​for the climb task segment are shown in Table 6 below;

[0206] Table 6 High-load interception values ​​for the climbing task segment

[0207] Task segment High load cutoff value / g Section 1 Climb 0.8439

[0208] (2) Low-load cutoff

[0209] The low-load deletion values ​​for the climb task segment are shown in Table 7 below;

[0210] Table 7 Low-load deletion values ​​for the climb task segment

[0211] Task segment Low load delete value / g Section 1 Climb 0.15

[0212] S5: Load spectrum compilation;

[0213] S51: 5×5 spectrum compilation;

[0214] Iteratively compile 5×5 spectra, compile 5×5 spectra for each task segment under Profile 1. The following example uses the climbing task segment of Profile 1 for illustration.

[0215] First, the number of flight types in the gust spectrum was determined. The number of flight types in the flight spectrum is shown in Table 8. Table 8: Number of flight types in a typical flight profile (1000 flights).

[0216] Serial Number Typical flight profile A B C D E 1 Section 1 1 16 74 288 622

[0217] The initial load level ratio should be set according to Table 3, as shown in Table 9 below;

[0218] Table 9 Initial Load Level Ratio

[0219] <![CDATA[Load level ratio R i > Gust of Wind <![CDATA[R1]]> 0.335 <![CDATA[R2]]> 0.24 <![CDATA[R3]]> 0.25 <![CDATA[R4]]> 0.175

[0220] The initial equivalent overload Δn at each stage is obtained. ydoi and the corresponding initial equivalent load cycle number N eqoi As shown in the following 10;

[0221] Table 10 Initial Equivalent Load and Number of Initial Equivalent Load Cycles

[0222] <![CDATA[Initial equivalent load Δn ydoi > <![CDATA[Initial equivalent load cycle number N eqoi > 0.8439 1 0.7276 18 0.5288 139 0.3582 1178 0.2107 5229

[0223] B is determined based on the initial equivalent load cycle number. ij As shown in Table 11 below;

[0224] Table 11 Number of times each load level occurs under each flight type (B) ij

[0225]

[0226]

[0227] The number of overtakes and the probability of overtake P corresponding to each flight type i As shown in Table 12 below;

[0228] Table 12. Number of Overtakes and Probability for Each Flight Type

[0229] Flight type Number of flights Cumulative number of flights Exceeding the limit Exceeding probability A 1 1 0.12 0.00012 B 16 17 8.14 0.00814 C 74 91 35.51 0.03551 D 288 379 233.21 0.23321 E 622 1001 615.94 0.61594

[0230] The final determined load levels for each level are shown in Table 13 below;

[0231] Table 13 Final determined load level ratios for each level

[0232] <![CDATA[Load level ratio R i > <![CDATA[R1]]> 0.205 <![CDATA[R2]]> 0.28 <![CDATA[R3]]> 0.24 <![CDATA[R4]]> 0.275

[0233] The final equivalent loads at each level are shown in Table 14 below;

[0234] Table 14 Final Determined Equivalent Loads at Each Level

[0235] <![CDATA[Equivalent load at each level Δn ydi > <![CDATA[Δn yd1 ]]> 0.8439 <![CDATA[Δn yd2 ]]> 0.5498 <![CDATA[Δn yd3 ]]> 0.4790 <![CDATA[Δn yd4 ]]> 0.3282 <![CDATA[Δn yd5 ]]> 0.2434

[0236] Using lgΔn ydi Calculation formula for fitting data pairs (lgΔn) ydi ,u p,i The correlation coefficient was 0.9984, and the resulting fitted curve is shown below. Figure 6 As shown;

[0237] The final determined 5×5 spectrum of the sudden wind overload in the climbing mission section under section 1 is shown in Table 15;

[0238] Table 15 Profile 1: 5×5 Spectrum of Sudden Winds in the Lower Climbing Section

[0239]

[0240] S52: Random compilation of load spectrum;

[0241] According to the method in S51, a fatigue load spectrum (load time series) was compiled, where the random spectrum of the climb mission segment of profile 1A flight type is as follows: Figure 7 As shown.

[0242] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0243] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A 5×5 spectrum iteration compilation method based on overload cumulative transcendence number curves, characterized in that, Includes the following steps: Step 1: Generate an aircraft mission profile based on the usage requirements of the aircraft under test, and determine the sequence, time ratio, parameters, and total number and ratio of aircraft mission profiles under a block spectrum. Step 2: Collect measured load data of the aircraft under test to generate overload exceedance curves and perform preprocessing; Step 3: Collect and count the peak and valley values ​​of the preprocessed overload exceedance number curve to obtain the cumulative overload exceedance number curve, calculate the load level ratio of the cumulative overload exceedance number curve for each task segment, and segment the curve according to the load level ratio. Step 4: Calculate the equivalent loads of each level of the cumulative overload exceedance curve for each mission segment and fit them to obtain the correlation coefficients. If the correlation coefficients are less than the set coefficient threshold, return to Step 3 to adjust the load level ratio; otherwise, obtain the 5×5 spectrum of each mission segment that constitutes the aircraft mission profile with the equivalent loads of each level. Step 5: Compile the mission segment spectrum based on the 5×5 spectrum of each mission segment, and compile the mission profile spectrum based on the mission segment spectrum and the aircraft mission profile. Step 6: Sort the mission profile spectrum and compile the flight-continued-flight spectrum; The measured load data includes gust load and maneuver load, generating overload exceedance curves. Preprocessing includes separating gust load and maneuver load and standard time accumulation processing. Through preprocessing, the task segment overload accumulation exceedance data for each load level, as well as maneuver load data and gust load data, are obtained. The process includes: Step 21: Convert the cumulative overload ... Step 22: The separation conditions for separating the motor load data and the gust load data are as follows: Overload change rate less than the set threshold and duration exceeding 2 seconds is considered motor load data; otherwise, it is considered gust load data. Overloads with an absolute value of 4° or greater on the control surface deflection are considered maneuver load data; otherwise, they are considered gust load data. The specific process of step 5 is as follows: Step 51: Based on the 5×5 spectrum of each mission segment, link the payloads at each level and their corresponding frequencies, randomly and alternately select peak and valley values ​​for random pairing, and compile payload pair sequences f for all flight types under all mission segments. i,j,k Obtain the mission segment spectrum F of the k-th flight type from the j-th mission profile. j,k ; Step 52: Based on the mission segment spectrum, sort the mission segments according to the order of the mission profile to obtain a complete flight mission profile spectrum f. 1,j,k ,f 2,j,k ,f 3,j,k ,f 4,j,k ...f p,j,k ; Step 53: Obtain the task profile spectrum of all types by sorting, and represent it in vector form as follows: 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 ], L represents the number of task segments under the task profile, and K represents the number of task profiles.

2. The method for iterative compilation of a 5×5 spectrum based on an overload cumulative transcendence number curve according to claim 1, characterized in that, The specific process of collecting and counting peak and valley values ​​of the preprocessed overload exceedance number curve for each mission segment of the aircraft mission profile to obtain the cumulative overload exceedance number curve for each mission segment, and segmenting the curve according to the load level ratio, is as follows: Step 31: Collect peak and valley values ​​for the separated gust load data and maneuver load data for each task segment to obtain the corresponding gust overload peak and valley value data and maneuver overload peak and valley value data for that task segment; Step 32: Use the restricted cross-peak counting method to count the time history of the two types of overload peak and valley data after separation to obtain the peak value and valley value; Step 33: Select several load levels, and for the gust load data and maneuver load data, calculate the cumulative exceedance number N(+Δn) of the positive flight overload corresponding to the peak value. y The negative flight overload cumulative exceedance number N(-Δn) corresponding to the valley value. y The average of the cumulative exceedance numbers corresponding to the peak and valley overload values ​​is calculated as the overload Δn. y The cumulative transcendence number N(Δn) y ), N(Δn y ) is represented as: Obtain the overload cumulative exceedance data pair (Δn) y N(Δn) y )) i , where i represents the i-th level load; Step 34: Standardize the cumulative exceedance number of the specified load to obtain the standard cumulative exceedance number N of the i-th level load at standard time. i , is represented as: Among them, t M t represents the flight time corresponding to the measured payload data. S N(Δn) represents the standard time for the task segment. yi ( ) represents the cumulative overload exceedance number of the i-th level load obtained from the measured time; step 34 is performed on the cumulative overload exceedance number of all loads to obtain the standard overload cumulative exceedance number data pair (Δn) of the i-th level load for all takeoffs and landings in each mission segment. y ,N) i ; Step 35: For each mission segment of the i-th load, the standard cumulative overload exceedance data pairs (Δn) for all takeoffs and landings. y ,N) i The cumulative overload exceedance number N for this mission segment under a single takeoff and landing is obtained by summing and averaging. t , is represented as: Where, N i,j N represents the standard cumulative exceedance number of the mission segment under the i-th level load during the j-th takeoff and landing; ti This represents the cumulative overload exceedance number for this mission segment under a single takeoff and landing of the i-th level load; j represents the takeoff and landing sequence number; m represents the total number of actual takeoffs and landings. Step 36: Obtain the cumulative overload exceedance data pair (Δn) for the i-th level load under a single takeoff and landing based on the cumulative overload exceedance number for this mission segment under a single takeoff and landing. y N t ) i For a single takeoff and landing under the i-th level load, the cumulative overload exceedance data pair (Δn) y N t ) i By performing fitting, the cumulative overload exceedance number curve for each task segment is obtained, expressed as: Δn y =a*lgN t +b a and b both represent the parameters of the linear fit; Step 37: Discretize the cumulative overload exceedance curve of each task segment into 4 segments, and represent them using the horizontal axis interval according to the low load cutoff value and the high load cutoff value, respectively represented as (LB5,LB4), (LB4,LB3), (LB3,LB2) and (LB2,LB1), where LB5 is the low load cutoff value, LB1 is the high load cutoff value, and LB2, LB3 and LB4 are the median values ​​of the low load cutoff value and the high load cutoff value, respectively; Step 38: Calculate the load level ratio R for each discrete curve segment. i The expression is: R i =(LB i -LB i+1 ) / (LB1-LB5) i＝1,2,3,4； Step 39: Replace each discrete curve segment with m straight lines. The linear equation of the i-th discrete curve segment is expressed as: Δn y =a i lgN t +b i Among them, a i b i All are constants of the i-th discrete curve; Δn yd N represents the equivalent load, and N is the average value of the x-axis interval of each segment of the discrete curve. eq The equivalent load cycle number is S; S is the slope parameter of the material's SN curve. Step 310: Obtain the equivalent load Δn of the four discrete curves corresponding to the four levels of wind load data and motor load data, respectively. ydi (i = 2, 3, 4, 5) and the corresponding number of load cycles N eqi (i = 2, 3, 4, 5); the highest load level LB5 (Δn) yd1 As a high load constituting a level 5 load, the corresponding number of load cycles is taken as the corresponding occurrence number.

3. The method for iterative compilation of a 5×5 spectrum based on an overload cumulative transcendence number curve according to claim 2, characterized in that, The specific process for compiling the 5×5 spectra for each task segment of the gust load data and maneuver load data in step 4 is as follows: Step 41: Determine the frequency y of each flight type i ; Step 42: Determine the initial equivalent load for each stage and the number of cycles for each initial equivalent load; Set an initial load level ratio to determine the starting point of the four discrete curves; Take the initial equivalent load Δn at each stage ydoi The average value of each discrete curve interval is used to calculate the number of cycles N of the initial equivalent load at each stage based on the initial equivalent load at each stage. eqoi ; Step 43: Determine the frequency of each load level under each flight type; Based on the initial equivalent load at each stage and the number of cycles of the initial equivalent load at each stage, and in accordance with the load similarity principle for each flight type, the number of occurrences of each load stage B under each flight type is determined. ij And the corresponding cumulative number of incremental overload cycles N at each level eqci ; Step 44: Analyze the occurrence frequency (y) of each flight type i The cumulative number of flights (BE) for each flight type is obtained by summing them up. i , is represented as: Step 45: Based on cumulative flight count (BE) i The geometric mean method is used to obtain the transcendental number degree (LDE). i , is represented as: LDE i =(BE i *THEY i-1 )^0.5; Step 46: Divide the number of transcendences by the total number of rises and falls N corresponding to the block spectrum. c The exceedance probability P corresponding to each load level is obtained. i The expression is: P i =LDE i / N c ; Step 47: Calculate the equivalent load Δn for each stage ydi The probability of exceeding P i and the corresponding standard normal distribution quantile u p,i The data pair (lgΔn) is obtained. ydi ,u p,i (i = 2, 3, 4, 5); data pairs (lgΔn) ydi ,u p,i The correlation coefficient r and lgΔn are obtained by fitting the data. ydi =μ+u p,i σ and μ represent the mean, and σ represents the standard deviation; Step 48: Determine whether the correlation coefficient is greater than the set coefficient threshold. If it is greater than the set coefficient threshold, output the 5×5 spectrum of each task segment; otherwise, return to step 38.

4. The method for iterative compilation of a 5×5 spectrum based on an overload cumulative transcendental number curve according to claim 1, characterized in that, The specific process of step 6 is as follows: Step 61: Categorize all types of mission profiles according to flight type; Step 62: Randomly arrange the flight-continued-flight patterns according to the frequency of occurrence of each flight type under the block spectrum to form the flight-continued-flight pattern.

5. The method for iterative compilation of a 5×5 spectrum based on an overload cumulative transcendental number curve according to claim 1, characterized in that, The principles for dividing task segments are as follows: The starting point of the climb is determined by the flap deflection angle becoming 0°. The end point of the climb is determined by the inflection point where the altitude curve flattens out; The end point of level flight is determined by the inflection point of the descent curve; The descent end point is determined by the flap deflection angle becoming 35°.

6. The method for iterative compilation of a 5×5 spectrum based on an overload cumulative transcendental number curve according to claim 2, characterized in that, The requirements for time history counting are: For positive overload, the 0g load state of each task segment is used as the baseline. When the valley value between two peaks or the peak value between two valleys does not exceed the upper deviation of the baseline of that task segment, the maximum peak value is recorded. For negative overload, the 0g load state of each task segment is used as the baseline. When the valley value between two peaks or the peak value between two valleys does not exceed the lower deviation of the baseline of the task segment, the maximum valley value is recorded. The upper and lower deviations are 20% of the maximum peak value.

Images