CN106096262A - A kind of aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation - Google Patents

Abstract

The invention discloses a kind of aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation, the loading spectrum of each parameter in different flight profile, mission profiles is used and adds up based on rain flow method；After filtering the load cycle that amplitude is less than amplitude thresholds, obtain the period of each parameter of each section, the different parameter loading spectrum of same flight profile, mission profile obtains different periods, different parameters is added up at each section the period one_to_one corresponding obtained, each load parameter can get string loop-around data, and then assumed (specified) load composes the correlation coefficient between each parameter.Invention removes the delay phenomenon between engine speed and overload values, result is more consistent with reality.The correlation coefficient of the normal g-load value largely representing motor-driven circulation and the rotating speed representing air circulation is close to 1.Foundation is provided with assessment further for aero-engine biometry.

Description

A kind of aero-engine loading spectrum Calculation of correlation factor based on rain-flow counting circulation Method

Technical field

The present invention relates to a kind of aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation, belong to Aero-engine loading spectrum technical field.

Background technology

China has compared the most empty survey and statistics to various engines flight mission profile a few days ago, it is thus achieved that The most useful use information.How to utilize these flight mission profiles to work out rational loading spectrum, the stress for electromotor divides Analysis and Life Design provide according to just becoming very distinct issues.The most domestic establishment ENGINE FLIGHT LOAD time spectrum often Only consider the change of a flight profile, mission profile parameter, the i.e. establishment of one-parameter spectrum.The establishment aspect of one-parameter spectrum is domestic have been done A large amount of intensive research work.But, the load that most of aeroengine components are born is a multicomponent load Spectrum, the stress etc. born such as bearing, densification device, the wheel rim of the turbine disk, not only relevant with rotating speed, and with temperature, pressure Relevant etc. factor.And install the thrust of stress on joint and electromotor and 3 directions flight overload factor (axial load factor NX, Normal g-load NY, lateral overload NZ) relevant, it is a part bearing multiparameter load.One-parameter compiles spectrum for some part May be largely effective, it is the most not enough that the part bearing complex load is then seemed.The place of aero-engine flight mission profile Actually one multiparameter spectrum problem of reason. effect and the phase thereof of flight mission profile multiparameter must be considered carrying out compiling time spectrum Impact between Hu.But the correlation coefficient method for solving between each parameter is the most immature to be further improved.Therefore, grind Study carefully the dependency relation of multiparameter loading spectrum and research multiparameter loading spectrum preparation method is had important theory significance and engineering should By value.

Aero-engine flight load spectrum be random, as blade, the dither loading spectrum of rotor-support-foundation system and send out Motivation rotating speed spectrum, flight speed spectrum etc..Wherein engine speed spectrum, electromotor center of gravity normal direction overload factor spectrum and flight speed The spectral pattern of degree spectrum is closely related with operating with of aircraft, such as aircraft engine speed section table when maneuvering flight Now for from the change repeatedly being up between slow train, form typical slow train-maximum-slow train circulation, electromotor center of gravity normal direction mistake Carry coefficient spectrum then from 1.0 increase to a certain maximum after return 1.0, corresponding to maneuvering flight action one by one.By to The analysis of the engine load spectrum that operation is relevant finds (such as accompanying drawing 4), and the appearance of normal g-load coefficient task segment turns with electromotor Speed change has bigger dependency.

But the correlation analysis of existing maneuver load and aerodynamic loading be by each load parameter as one completely with Machine series calculates correlation coefficient, finds in conjunction with task segment analysis, and the initial time of each engine speed task segment is always Prior to the initial time of normal g-load task segment, time difference between the two did not waited (such as accompanying drawing 2) in several seconds to tens seconds, sent out The time to peak of motivation rotating speed task segment is (such as the accompanying drawing 3) of the time to peak prior to normal g-load task segment equally, according to upper The maneuver load that the method for stating is calculated is less than normal with the correlation coefficient of aerodynamic loading.It is thus desirable to existing loading spectrum parameter phase Close coefficient method for solving to make further improvements.

Summary of the invention

The present invention is directed to the deficiency of the problems referred to above, propose a kind of aero-engine loading spectrum phase based on rain-flow counting circulation Closing coefficient calculation method, the method can make up the machine that existing aero-engine loading spectrum Calculation of correlation factor method calculates The problem that dynamic loading is less than normal with the correlation coefficient of aerodynamic loading, and eliminate the delay phenomenon between rotating speed and overload values.

The present invention solves that the technical scheme that above-mentioned technical problem proposes is:

A kind of aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation, to different flight profile, mission profiles In each parameter loading spectrum use add up based on rain flow method.Provide the amplitude threshold of each parameter in different flight profile, mission profile Value, after filtering the load cycle that amplitude is less than amplitude thresholds, it is thus achieved that the period of each parameter of each section, same flight profile, mission profile is different Parameter loading spectrum obtain different periods, different parameters is added up, at each section, the period one_to_one corresponding that obtains, often Individual load parameter all can get string loop-around data, and then assumed (specified) load composes the correlation coefficient between each parameter.

Described comprise the following steps:

Step 1, obtains the number of the flight mission profile of aero-engine and the flight ginseng of each flight mission profile Number.

Step 2, carries out carrying of flight profile, mission profile peak-to-valley value to the flight parameter of each flight mission profile that step 1 obtains Take.

Step 3, the flight obtained the flight parameter of each flight mission profile obtained in step 1 according to step 2 is cutd open Face peak-to-valley value carries out rain flow method statistics.

Step 4, rejects each parameter amplitude less than [the 10% of minimum amplitude+(maximum amplitude-minimum amplitude)], then by width Value is rejected less than the normal g-load NY of threshold value 0.2, axial acceleration NX, lateral overload NZ, it is thus achieved that delete each ginseng after small magnitude Number amplitude number i.e. load cycle number, completes the removal of inefficient cycle amplitude.The removal of invalid amplitude can be after rain-flow counting Carry out, it is possible to carry out in rain-flow counting.

Step 5, obtains the parameters of each section through rain according to the flight parameter of the inefficient cycle amplitude of step 4 removal The number of the period after flow accounting, i.e. amplitude or average.

Step 6, each section that the flight parameter of each flight mission profile step 1 obtained obtains in step 5 each Correlation coefficient between individual parameter one_to_one corresponding, and then the assumed (specified) load each parameter of spectrum.

Preferred: the extracting method of flight profile, mission profile peak-to-valley value in described step 2: extract the every of each flight mission profile The peak-to-valley value of one flight parameter.

Preferred: in described step 3 before rain flow method is added up, first flight profile, mission profile peak-to-valley value is carried out series tune Whole, described method of adjustment: first the waveform that the flight parameter of each flight mission profile obtained in step 1 is formed is sealed Close process so that it is numerical value is connected from beginning to end.Again by waveform from the flight profile, mission profile peak-to-valley value that step 2 obtains peak-peak punishment Open so that it is join end to end and constitute new waveform.

Preferably, the statistical method of rain flow method in described step 3:

Step 31, the peak-peak of each flight parameter of each flight mission profile or minimum valley are starting point, Rearrange stress-time history.

Step 32, the rain stream originating in peak value will fall at next valley, or the rain stream originating in valley will be Falling at next peak value, the rain stream fallen is met following two kinds of situations and is then stopped: (a) originate in the rain stream of peak value run into equal to or Just stop higher than its peak value.B () originates in the rain stream of valley and runs into and just stop equal to or less than its valley.

Step 33, originates in peak value, the rain stream of valley runs into top-down rain stream and just stops, now according to this rain stream Beginning and end keeps a circulation in mind.

Step 34, takes out all of complete alternation, and records respective amplitude and average.

Preferred: the correlation coefficient that the assumed (specified) load in described step 6 is composed between each parameter is counted by below equation Calculate:

$r_{jk}=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\left(x_{ij}-\stackrel{\‾}{x_j}\right)\left(x_{ik}-\stackrel{\‾}{x_k}\right)}{\sqrt{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left(x_{ij}-\stackrel{\‾}{x_j}\right)}^2\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left(x_{ik}-\stackrel{\‾}{x_k}\right)}^2}}=\frac{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}{x}_{ik}-\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ik}}{\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_{ij}}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}\right)}^2}\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_{ik}}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ik}\right)}^2}}$

$\stackrel{\‾}{x_j}=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}{x}_{ij},\stackrel{\‾}{x_k}=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}{x}_{ik};$

Wherein, r_jkJth column data and the correlation coefficient of kth column data, x for matrix X_ijPass through for each parameter of each section The period of gained after rain-flow counting, i=1,2 ..., n；J=1,2 ..., p, p represent the number of load parameter, and n represents and cuts open Face number, X represents x_ijThe matrix formed.

The aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation of the present invention, compares existing skill Art, has the advantages that

The present invention propose based on rain-flow counting circulation aero-engine loading spectrum Calculation of correlation factor method, be based on Obtain to using relevant loading spectrum analysis, and eliminate the delay phenomenon between rotating speed and overload values, and by a large amount of outer number of fields According to verifying, the correlation coefficient of the normal g-load value and the rotating speed representing air circulation that largely represent motor-driven circulation connects Being bordering on 1, result is more consistent with reality, therefore compensate for existing aero-engine loading spectrum Calculation of correlation factor method and calculates The correlation coefficient of maneuver load and aerodynamic loading problem less than normal.Therefore the present invention can be aero-engine multiparameter loading spectrum Establishment, further for aero-engine biometry with assessment provide foundation.

Accompanying drawing explanation

Fig. 1 mono-section typical maneuvering flight loading spectrum segment.

Fig. 2 aero-engine rotating speed and the dependency relation of normal g-load initial time.

Fig. 3 aero-engine rotating speed and the dependency relation of normal g-load time to peak.

Dependency relation between Fig. 4 aero-engine rotating speed and normal g-load.

Fig. 5 rain flow method.

The computer programme flow diagram of Fig. 6 tri-peak-to-valley value rain-flow counting.

The normal g-load spectrum of Fig. 7 certain flight profile, mission profile of h type engine h.

The aero-engine loading spectrum Calculation of correlation factor method flow diagram that Fig. 8 circulates based on rain-flow counting.

Detailed description of the invention

Accompanying drawing discloses the structural representation of a preferred embodiment of the invention without limitation, detailed below with reference to accompanying drawing Technical scheme carefully is described.

Embodiment

The maneuvering flight loading spectrum of aero-engine is aircraft aeroplane engine in completing a maneuvering flight task process Machine time dependent curve loaded.Once typical maneuvering flight task is generally by several maneuvering flight action group Become, and each maneuvering flight action will necessarily make aero-engine generation one or many significant change loaded.Therefore, The maneuvering flight loading spectrum of aero-engine just can be described as being had by several amplitudes the load segment of significant change and form Load-time graph.Herein these load segments are defined as maneuvering flight load task section, are called for short task segment.Accompanying drawing 1 It it is one section of typical maneuvering flight loading spectrum segment.From figure 1 it appears that this maneuvering flight loading spectrum segment is by four differences Task segment composition.I-th task segment is from moment T_iStart, elapsed time D_iRear end；I+1 task segment is from moment T_i+1Open Begin, elapsed time D_i+1Rear end.In Fig. 1, D_iIt is the persistent period of i-th task segment, T_iWhen being the arrival of i-th task segment Between, A_iIt is the peak value of i-th task segment, W_iIt is the waiting time between i-th task segment and i+1 task segment, T_i+1-T_iIt is Time interval between i+1 the task segment time of advent and i-th task segment time of advent.

Finding in conjunction with task segment analysis, as shown in Figure 4, the appearance of normal g-load coefficient task segment becomes with engine speed Change and there is bigger dependency.But existing maneuver load is to be worked as by each load parameter with the correlation analysis of aerodynamic loading Making a completely random series and calculate correlation coefficient, the maneuver load calculated is less than normal with the correlation coefficient of aerodynamic loading.

In order to make up maneuver load that existing aero-engine loading spectrum Calculation of correlation factor method calculates with pneumatic The problem that the correlation coefficient of load is less than normal, the present invention proposes aero-engine loading spectrum phase relation based on rain-flow counting circulation Number calculating method, is to obtain based on to using relevant loading spectrum analysis, and it is existing to eliminate the delay between rotating speed and overload values As, and verified by a large amount of field datas, result is more consistent with practical situation.Thus carry out maneuver load and aerodynamic loading Dependency relation research and the establishment of aero-engine multiparameter loading spectrum, carry with assessment for aero-engine biometry further For foundation.

A kind of aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation, as shown in Figure 8, to not Use to enter based on rain flow method with the loading spectrum (such as rotating speed spectrum, center of gravity normal direction overload factor spectrum etc.) of parameter each in flight profile, mission profile Row statistics.Provide the amplitude thresholds of each parameter in different flight profile, mission profile, after filtering the load cycle that amplitude is less than amplitude thresholds, obtain The period of each parameter of each section, the different parameter loading spectrum of same flight profile, mission profile obtains different periods, difference is joined Number adds up the period one_to_one corresponding (as shown in table 3) obtained at each section, and each load parameter can get string circulation Correlation coefficient between data, and then the assumed (specified) load each parameter of spectrum.Rain flow way of the present invention is that a kind of conventional process carries The two-parameter cycle count method of lotus spectrum time history, by two parameters in statistics load cycle, reflection load cycle is several Whole information.

Specifically include following steps:

Step 1, obtains the number of the flight mission profile of aero-engine and the flight ginseng of each flight mission profile Number.

If a total of n flight mission profile, each flight mission profile has m flight parameter, such as aero-engine Normal g-load Ny, lateral overload Nz, axial acceleration Nx, high pressure rotor rotating speed n2_L etc. is sent out on an electromotor left side.Each flight profile, mission profile is each Flight parameter has numerous flying quality.

Step 2, the extraction of flight profile, mission profile peak-to-valley value: the flight parameter of each flight mission profile that step 1 obtains is entered The extraction of row flight profile, mission profile peak-to-valley value.

Wherein, the extracting method of flight profile, mission profile peak-to-valley value: extract each flight parameter of each flight mission profile Peak-to-valley value.

The essence that peak-to-valley value extracts is exactly to disregard the time factor of cyclic loading, by peak (paddy) value of handled parameter, presses Time sequencing extracts from flight profile, mission profile, and forms the peak-to-valley value sequence being sequentially arranged, and has greatly reduced section Data, it is simple to carry out parameter cyclic counting.

Step 3, parameter cyclic counts: to the flight parameter of each flight mission profile obtained in step 1 according to step 2 The flight profile, mission profile peak-to-valley value obtained carries out rain flow method statistics.

This supplemental characteristic of this section is carried out rain-flow counting.Before rain-flow counting, peak-to-valley value series adjustment to be carried out, To avoid the rain stream later stage waveform of divergent-convergent type occur, as shown in Figure 5, its method is first each by obtain in step 1 The waveform that the flight parameter of flight mission profile is formed carries out sealing treatment so that it is numerical value is connected from beginning to end.Again by waveform from step At peak-peak in the rapid 2 flight profile, mission profile peak-to-valley values obtained separately so that it is join end to end and constitute new waveform (shown in Fig. 5), Then rain-flow counting is carried out.

The statistical method of rain flow method:

Step 31, the peak-peak of each flight parameter of each flight mission profile or minimum valley are starting point, Rearrange stress-time history.

Step 32, the rain stream originating in peak value will fall at next valley, or the rain stream originating in valley will be Falling at next peak value, the rain stream fallen is met following two kinds of situations and is then stopped: (a) originate in the rain stream of peak value run into equal to or Just stop higher than its peak value.B () originates in the rain stream of valley and runs into and just stop equal to or less than its valley.

Step 33, originates in peak value, the rain stream of valley runs into top-down rain stream and just stops, now according to this rain stream Beginning and end keeps a circulation in mind.

Step 34, takes out all of complete alternation, and records respective amplitude and average.

Specific program should be worked out and be automatically performed rain-flow counting process by computer.The rain-flow counting of load cycle is had many Planting programmed method, (accompanying drawing 6) is the computer programme flow diagram using three peak-to-valley value rain-flow countings.

Step 4, the removal of inefficient cycle amplitude: by each parameter amplitude less than [minimum amplitude+(maximum amplitude-the most small size Value) 10%] reject, then by amplitude less than the normal g-load NY of threshold value 0.2, axial acceleration NX, lateral overload NZ reject, obtain The each i.e. load cycle number of parameter amplitude number after small magnitude must be deleted, complete the removal of inefficient cycle amplitude.Invalid amplitude Removal can be carried out after rain-flow counting, it is possible to carries out in rain-flow counting.

Step 5, the statistical counting of circulation: according to the flight parameter each section of acquisition of the inefficient cycle amplitude that step 4 is removed Parameters period after rain-flow counting, i.e. amplitude or the number of average.

If a total of p load parameter of loading spectrum, section number is n, each section each parameter gained after rain-flow counting Period be x_ij(i=1,2 ..., n；J=1,2 ..., p), then raw data matrix is:

$X=\left[\begin{array}{cccc}{x}_{11}& x_{12}& ...& x_{1p}\\ x_{21}& x_{22}& ...& x_{2p}\\ ......& & & \\ x_{n1}& x_{n2}& ...& x_{np}\end{array}\right].$

Step 6, each section that the flight parameter of each flight mission profile step 1 obtained obtains in step 5 each Correlation coefficient between individual parameter one_to_one corresponding (as shown in table 3), and then the assumed (specified) load each parameter of spectrum.

The correlation matrix R being calculated between each parameter:

$R=\left[\begin{array}{cccc}{r}_{11}& r_{12}& ...& r_{1p}\\ r_{21}& r_{22}& ...& r_{2p}\\ \begin{array}{c}.\\ .\\ .\end{array}& \begin{array}{c}.\\ .\\ .\end{array}& \begin{array}{c}.\\ .\\ .\end{array}& \begin{array}{c}.\\ .\\ .\end{array}\\ r_{p1}& r_{p2}& ...& r_{pp}\end{array}\right]$

Wherein: r_jkFor jth column data and the correlation coefficient of kth column data of matrix X,

$r_{jk}=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\left(x_{ij}-\stackrel{\‾}{x_j}\right)\left(x_{ik}-\stackrel{\‾}{x_k}\right)}{\sqrt{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left(x_{ij}-\stackrel{\‾}{x_j}\right)}^2\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left(x_{ik}-\stackrel{\‾}{x_k}\right)}^2}}=\frac{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}{x}_{ik}-\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ik}}{\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_{ij}}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}\right)}^2}\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_{ik}}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ik}\right)}^2}}$

$\stackrel{\‾}{x_j}=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}{x}_{ij},\stackrel{\‾}{x_k}=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}{x}_{ik};$

Wherein, r_jkJth column data and the correlation coefficient of kth column data, x for matrix X_ijPass through for each parameter of each section The period of gained after rain-flow counting, i=1,2 ..., n；J=1,2 ..., p, p represent the number of load parameter, and n represents and cuts open Face number, X represents x_ijThe matrix formed.

(1) statistical result of certain h type engine h dozens of flight profile, mission profile:

Having added up 84 flight profile, mission profiles altogether, each flight profile, mission profile has added up 10 parameters, including high pressure rotor rotating speed N2-L, Normal g-load Ny, lateral overload Nz, axial acceleration Nx, pitch rate FYJSL, rolling angle rate HGJSL, course angle speed HXJSL, X-axis acceleration A x, Y-axis acceleration A y, Z axis acceleration A z；Accompanying drawing 7 is the normal g-load Ny spectrum of certain flight profile, mission profile, Totally 6167 sampled points, sample frequency is 1HZ.

(1) extraction of flight profile, mission profile peak-to-valley value.Extract the peak-to-valley value of certain certain parameter of section, the essence that peak-to-valley value extracts It is exactly the time factor disregarding cyclic loading, by peak (paddy) value of handled parameter, extracts from flight profile, mission profile in chronological order Out, and form the peak-to-valley value sequence being sequentially arranged, greatly reduce cross-sectional data, it is simple to carry out parameter cyclic meter Number.After extraction, each supplemental characteristic has been reduced to 2717 by original 6167, has greatly reduced cross-sectional data.

(2) parameter cyclic counting.This parameter loading spectrum of this section is carried out rain-flow counting, before rain-flow counting, first wants Carry out peak-to-valley value series adjustment, to avoid the rain stream later stage that the waveform of divergent-convergent type occurs.Its method is first to be sealed by waveform Close process so that it is numerical value is connected from beginning to end；Again by waveform at peak-peak separately so that it is join end to end constitute new waveform (as Accompanying drawing 5), then carry out rain-flow counting.

The counting rule of rain flow way is:

A) stress-time history is rearranged, with peak-peak or minimum valley as starting point；

B) the rain stream originating in peak value (valley) will fall at next valley (peak value) place, and the rain stream fallen meets following two Kind of situation then stops: (a) originates in the rain stream of peak value and run into and just stop equal to or higher than its peak value；

B () originates in the rain stream of valley and runs into and just stop equal to or less than its valley；

C) originate in peak value, the rain stream of valley runs into top-down rain stream and just stops, now can rising according to this rain stream Point and terminal keep a circulation in mind；

D) take out all of complete alternation, and record respective amplitude and average.

Specific program should be worked out and be automatically performed rain-flow counting process by computer.The rain-flow counting of load cycle is had many Planting programmed method, (accompanying drawing 6) is the computer programme flow diagram using three peak-to-valley value rain-flow countings.The peak-to-valley value extracted is carried out rain Flow accounting, the period finally given is 1358, and minimum amplitude is 0.005, and maximum amplitude is 1.3265.

(3) removal of inefficient cycle amplitude.The cyclic fatigue damage of component depends primarily on the size of circulation width.From peak valley Value sequence removes some fatigue damage is affected little a small amount of circulate in and be very important in engineering.By each parameter amplitude Reject less than [the 10% of minimum amplitude+(maximum amplitude-minimum amplitude)], then amplitude is less than the normal g-load of threshold value 0.2 NY, axial acceleration NX, lateral overload NZ reject, it is thus achieved that delete the number of each i.e. amplitude of parameter cyclic number after small magnitude.Invalid The place to go of amplitude can be carried out after rain-flow counting, it is possible to carries out in rain-flow counting.

Result after certain flight profile, mission profile normal g-load value rain-flow counting of table 1

(4) statistical counting circulated.Obtaining this parameter loading spectrum period after rain-flow counting of this section is 37.

(5) period after other parameters of this section count is obtained based on rain flow way.With reference to said method, can by programming The each supplemental characteristic of this section of batch processing, it is thus achieved that this section each parameter period after rain-flow counting.

Result after certain flight profile, mission profile of table 2 each parameter rain-flow counting

Numbering	N2L	Ny	Nz	Nx	FYJSL	HGJSL	HXJSL	Ax	Ay	Az
											12	15	37	0	3	107	48	8	4	1	12

(6) each parameter of other sections is obtained based on the period after rain flow way.With reference to said method, can batch by programming Process each supplemental characteristic of other sections, it is thus achieved that other sections each parameter period after rain-flow counting.

Result after more than 3 flight profile, mission profile each parameter rain-flow counting of table

(7) correlation coefficient between each parameter is calculated.System according to each parameter of each section based on rain stream statistics in upper table Meter result, can obtain the such as following table of the correlation matrix between each parameter by formula.

$r_{jk}=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\left(x_{ij}-\stackrel{\‾}{x_j}\right)\left(x_{ik}-\stackrel{\‾}{x_k}\right)}{\sqrt{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left(x_{ij}-\stackrel{\‾}{x_j}\right)}^2\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left(x_{ik}-\stackrel{\‾}{x_k}\right)}^2}}=\frac{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}{x}_{ik}-\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ik}}{\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_{ij}}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}\right)}^2}\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_{ik}}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ik}\right)}^2}}$

The correlation coefficient of the loading spectrum parameter that table 4 obtains based on rain-flow counting

(8) compare correlation coefficient and the difference of the correlation coefficient calculated based on rain flow way that traditional method calculates, enter Row is analyzed.Traditional method is to be stitched together by all sections, adds up the correlation coefficient between each parameter.

The correlation coefficient of table 5 loading spectrum parameter based on traditional Calculation of correlation factor method

Compared with traditional Calculation of correlation factor method, Calculation of correlation factor methods and results based on rain-flow counting circulation shows Show:

(1) between the parameter having, correlation coefficient has increased, and between some coefficients, correlation coefficient has reduced.Such as: main generation Correlation coefficient between the normal g-load Ny and rotating speed N2L of table maneuver load and aerodynamic loading dependency relation is increased by 0.28156 It is 0.90617.The correlation coefficient of N2l Yu Nx is increased to 0.87458 by 0.44401 the most respectively.Correlation coefficient between Ay and Ny It is kept to 0.73446 by 0.96594.

(2) represent correlation coefficient between the engine speed of gas force size and maneuver load and have significantly increase, especially It is that the correlation coefficient between normal g-load Ny and rotating speed N2L is increased to 0.90617 by 0.28156.Reject overload values and rotating speed it Between phase contrast factor after, the result obtained obviously more is consistent with practical situation.

(3) between partial parameters, correlation coefficient is the least, and substantially close to 0, this shows between these parameters the most permissible Regard as separate.Such as the correlation coefficient between lateral overload Nz and X-axis acceleration A x only-0.01175.

Visible by contrast, aero-engine loading spectrum correlation coefficient meter based on rain-flow counting circulation of the present invention Calculation method, is to obtain based on to using relevant loading spectrum analysis, and eliminates the delay phenomenon between rotating speed and overload values, and Verified by a large amount of field datas, largely represent the normal g-load value of motor-driven circulation and the rotating speed representing air circulation Correlation coefficient close to 1.Result is more consistent with reality..

It is merely to illustrate embodiments of the present invention above in conjunction with the preferred embodiment of the present invention described by accompanying drawing, and Not as to aforementioned invention purpose and claims content and the restriction of scope, every technical spirit according to the present invention To any simple modification made for any of the above embodiments, equivalent variations and modification, the most still belong to the technology of the present invention and rights protection category.

Claims (6)

1. an aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation, it is characterised in that: to not Use with the loading spectrum of parameter each in flight profile, mission profile and add up based on rain flow method；Provide each parameter in different flight profile, mission profile Amplitude thresholds, filter amplitude less than amplitude thresholds load cycle after, it is thus achieved that the period of each parameter of each section, same flight The different parameter loading spectrum of section obtains different periods, and the period obtained in each section statistics by different parameters is one by one Correspondence, each load parameter can get string loop-around data, and then assumed (specified) load composes the correlation coefficient between each parameter.

Aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation the most according to claim 1, It is characterized in that, comprise the following steps:

Step 1, obtains number and the flight parameter of each flight mission profile of the flight mission profile of aero-engine；

Step 2, the flight parameter of each flight mission profile obtaining step 1 carries out the extraction of flight profile, mission profile peak-to-valley value；

Step 3, the flight profile, mission profile peak that the flight parameter of each flight mission profile obtained in step 1 is obtained according to step 2 Valley carries out rain flow method statistics；

Step 4, rejects each parameter amplitude less than [the 10% of minimum amplitude+(maximum amplitude-minimum amplitude)], then amplitude is little Reject in normal g-load NY, axial acceleration NX, the lateral overload NZ of threshold value 0.2, it is thus achieved that delete each parameter width after small magnitude Value number i.e. load cycle number, completes the removal of inefficient cycle amplitude；The removal of invalid amplitude can be carried out after rain-flow counting, Also can carry out in rain-flow counting；

Step 5, obtains the parameters of each section through rain flowmeter according to the flight parameter of the inefficient cycle amplitude of step 4 removal The number of the period after number, i.e. amplitude or average；

Step 6, each ginseng of each section that the flight parameter of each flight mission profile step 1 obtained obtains in step 5 Correlation coefficient between number one_to_one corresponding, and then the assumed (specified) load each parameter of spectrum.

Aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation the most according to claim 2, It is characterized in that: the extracting method of flight profile, mission profile peak-to-valley value in described step 2: extract each of each flight mission profile The peak-to-valley value of flight parameter.

Aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation the most according to claim 2, It is characterized in that: in described step 3 before rain flow method is added up, first flight profile, mission profile peak-to-valley value is carried out series adjustment, Described method of adjustment: first the waveform that the flight parameter of each flight mission profile obtained in step 1 is formed is closed Process so that it is numerical value is connected from beginning to end；Again by waveform peak-peak from the flight profile, mission profile peak-to-valley value that step 2 obtains separately, Make it join end to end and constitute new waveform.

Aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation the most according to claim 2, It is characterized in that, the statistical method of rain flow method in described step 3:

Step 31, the peak-peak of each flight parameter of each flight mission profile or minimum valley are starting point, again Arrange stress-time history；

Step 32, the rain stream originating in peak value will fall at next valley, or the rain stream originating in valley will be at next Falling at individual peak value, the rain stream that falls is met following two kinds of situations and is then stopped: (a) originates in the rain stream of peak value and run into and equal to or higher than Its peak value just stops；B () originates in the rain stream of valley and runs into and just stop equal to or less than its valley；

Step 33, originates in peak value, the rain stream of valley runs into top-down rain stream and just stops, now according to the starting point of this rain stream A circulation is kept in mind with terminal；

Step 34, takes out all of complete alternation, and records respective amplitude and average.

Aero-engine loading spectrum Calculation of correlation factor method based on rain-flow counting circulation the most according to claim 2, It is characterized in that: the correlation coefficient that the assumed (specified) load in described step 6 is composed between each parameter is calculated by below equation:

$r_{jk}=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\left(x_{ij}-\stackrel{\‾}{x_j}\right)\left(x_{ik}-\stackrel{\‾}{x_k}\right)}{\sqrt{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left(x_{ij}-\stackrel{\‾}{x_j}\right)}^2\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left(x_{ik}-\stackrel{\‾}{x_k}\right)}^2}}=\frac{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}{x}_{ik}-\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ik}}{\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_{ij}}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ij}\right)}^2}\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_{ik}}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_{ik}\right)}^2}}$

$\stackrel{\‾}{x_j}=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}{x}_{ij},\stackrel{\‾}{x_k}=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}{x}_{ik};$

Wherein, r_jkJth column data and the correlation coefficient of kth column data, x for matrix X_ijFor each parameter of each section through rain flowmeter The period of gained after number, i=1,2 ..., n；J=1,2 ..., p, p represent the number of load parameter, and n represents section number, X represents x_ijThe matrix formed.