CN103473470B - A kind of ground effect wind tunnel test data processing method - Google Patents

Abstract

A kind of ground effect wind tunnel test data processing method, the present invention utilizes and is applicable to the data correction formula of ground effect and coefficient regression solves the wind tunnel test data that band can quickly, objectively disturb by computational methods and carries out the correction of system；Test data is brought into and is obtained coefficient in formula in correction formula by this method, utilizes regression algorithm to obtain a system number the highest with initial data degree of association as correction formula coefficient, then utilizes the formula determined to calculate the data identical with testing site state.The result of this method shows that revised data have the highest reduction degree, it is possible to for the basic data that the post processing of data is good with analyzing offer rule, create good condition for ground effect vehicle profile type selecting and profile Fine design.

Description

A kind of ground effect wind tunnel test data processing method

Technical field

The present invention relates to the post-processing approach of a kind of ground effect wind tunnel test data, for the near-earth stability of analytically effect aircraft, belong to aircraft test and technical field of measurement and test.

Background technology

General ground effect testing is exploratory flight device aerodynamic characteristics on the differing heights of ground, therefore its aerodynamic data variable parameter has more height variable h compared with conventional aircraft, various model attitude angle under differing heights to be carried out and angle of rudder reflection test in test, test data is except conventional c_L～α, m_z～also include outside αData above is used for calculating " angle of attack focus " and " flying high focus ".According to definition,

Angle of attack focus: $\stackrel{\‾}{X_{\mathrm{f\α}}}=\stackrel{\‾}{X_t}-m_Z^{\mathrm{\α}}/C_L^{\mathrm{\α}}$ Fly high focus: $\stackrel{\‾}{X_{\mathrm{fh}}}=\stackrel{\‾}{X_t}-m_Z^h/C_L^h$

According to near-earth static-stability condition:

Visible, test data c_L～α, m_z～α,Precision determine the computational accuracy of two focus datas.Routine test (without ground effect) technology maturation, interference factor are little and are prone to eliminate, and are generally possible to obtain the flat curve that precision is higher；And after adding ground boundary condition, due to the restriction of experimental technique cannot truly simulate floor and flight relative motion and boundary region situation, therefore test is disturbed more serious, obtainsOften there is singular point and curve smooth degree bad (distortion) in curve, can have a strong impact onSolving precision, causing " flying high focus " curve to fluctuate more greatly cannot accurate evaluation near-earth stability.

Summary of the invention

The technology of the present invention solves problem: overcome the deficiencies in the prior art, it is provided that the processing method of a kind of ground effect wind tunnel test data, eliminates the test interference of test data analyzer, provides condition for stability Accurate Analysis.

The technical solution of the present invention is:

A kind of ground effect wind tunnel test data processing method, comprises the following steps that

(1) obtain ground effect wind tunnel test data and transform it into the data form relevant with height

(2) test data after conversion in step (1) is carried out permutation and combination and forms multi-group data combination, often the combination of group data comprises 3 test data points；

(3) the test data point often organized in data combination substituting into formula (1) respectively and solve following formula coefficient, multi-group data combination can obtain to organize corresponding coefficient more:

$(C_y+a)\×(\stackrel{\‾}{h}+b)=C$

$(m_Z+a_1)\×(\stackrel{\‾}{h}+b_1)=C_1---\left(1\right)$

Wherein, C_yFor lift coefficient, a, b, C are lift formula generation to ask coefficient, m_ZPitching moment coefficient is, a₁、b₁、C₁Coefficient is sought for pitching moment formula generation,For height coefficient；(4) judge whether the combination of the multi-group data in step (2) completes solving in step (3), if completing, entering step (5), otherwise entering step (3)；

(5) the many groups equation coefficients solved in step (3) being grouped, according to often organizing packet containing 1 system number, 2 system numbers .... many systems number is grouped, and average coefficient value is asked in the combination to including many systems number, is normalized to 1 system number；

(6) each system number solved in step (5) is substituted into formula (1) and carry out regression Calculation with original experiment data respectively, try to achieve the multiple correlation coefficient of formula；

(7) judge whether the combination of the many groups equation coefficients in step (5) completes to solve multiple correlation coefficient in step (6), if completing, entering step (8), otherwise entering step (6)；

(8) one group of multiple correlation coefficient maximum is selected as test data correction equation coefficients；

(9) utilize the test data under the correction formula correction corresponding states that step (8) obtains, revise the test data obtaining under another state after terminating and enter step (1).

The present invention compared with prior art provides the benefit that:

(1) data are not modified relative to original technology or are only finely adjusted indivedual singular points (other data do not adjust) by the present invention, it is ensured that the systematicness of the data after adjustment, improve the stability that whole data process.

(2) present invention relative to original technology manually realize adjust, the present invention forms data correcting method and the program of complete set, it is possible to accomplish objective, be rapidly completed data correction.

(3) data correction mode of the present invention is simple, directly perceived, is more suitable for ground effect data relative to high-order interferential loads technology such as Fourier transformations and processes.

Accompanying drawing explanation

Fig. 1 is the inventive method flow chart；

Fig. 2 is that lift is with height change trial curve；

Fig. 3 is that moment is with height change trial curve；

Fig. 4 is that lift is with angle of attack variation trial curve；

Fig. 5 is that moment is with angle of attack variation trial curve；

Fig. 6 is that lift is with height change matched curve；

Fig. 7 is that moment is with height change matched curve；

Fig. 8 is that lift is with angle of attack variation matched curve；

Fig. 9 is that moment is with angle of attack variation matched curve；

Figure 10 is that lift is with height change correlation curve；

Figure 11 is that moment is with height change correlation curve；

Figure 12 is that lift is with angle of attack variation correlation curve；

Figure 13 is that moment is with angle of attack variation correlation curve；

Figure 14 is original focal figure；

Figure 15 is matching focus chart；

Figure 16 is original focal figure；

Figure 17 is matching focus chart.

Detailed description of the invention

Below in conjunction with the accompanying drawings the detailed description of the invention of the present invention is further described in detail.

As shown in Figure 1, a kind of ground effect wind tunnel test data processing method, the present invention is to be fitted processing by the data comprising interference that test obtains, and obtains the curve that the rule change based on test data smoothes out, and designing for follow-up stability analysis and control rate provides advantage.The present invention comprises the following steps that

(1) obtain ground effect wind tunnel test data and transform it into the data form relevant with heightWherein m_ZFor pitching moment coefficient,For height coefficient；

(2) test data after conversion in step (1) is carried out permutation and combination and forms multi-group data combination, often the combination of group data comprises 3 test data points；

(3) the test data point often organized in data combination substituting into formula (1) respectively and solve following formula coefficient, multi-group data combination can obtain to organize corresponding coefficient more:

$(C_L+a)\×(\stackrel{\‾}{h}+b)=C$

$(m_Z+a_1)\×(\stackrel{\‾}{h}+b_1)=C_1---\left(1\right)$

Wherein, C_LFor lift coefficient, a, b, C are lift formula generation to ask coefficient, m_ZPitching moment coefficient is, a₁、b₁、C₁Coefficient is sought for pitching moment formula generation,For height coefficient；

(4) judge whether the combination of the multi-group data in step (2) completes solving in step (3), if completing, entering step (5), otherwise entering step (3)；

(5) the many groups equation coefficients solved in step (3) being grouped, according to often organizing packet containing 1 system number, 2 system numbers .... many systems number is grouped, and average coefficient value is asked in the combination to including many systems number, is normalized to 1 system number；

(6) each system number solved in step (5) is substituted into formula (1) and carry out regression Calculation with original experiment data respectively, try to achieve the multiple correlation coefficient of formula；

(7) judge whether the combination of the many groups equation coefficients in step (5) completes to solve multiple correlation coefficient in step (6), if completing, entering step (8), otherwise entering step (6)；

(8) one group of multiple correlation coefficient maximum is selected as test data correction equation coefficients；

(9) utilize the test data under the correction formula correction corresponding states that step (8) obtains, revise the test data obtaining under another state after terminating and enter step (1).

3 groups of data just can obtain the coefficient (a, b, c) in one group of fitting formula in theory, therefore can obtain to organize coefficient in formula by original experiment data more, the all coefficients that will obtain, bring formula into including the meansigma methods of single system number and some system numbers and carry out regression Calculation, taking a system number the highest with initial data multiple correlation coefficient is fitting formula coefficient, is used for calculating data after correction.

The work process of the present invention is further illustrated below with an instantiation.For certain ground effect vehicle result of the test, the focus graph of a relation (α≤4 °) in the range of Low Angle Of Attack drawn according to the experimental data processing phenomenon that angle of attack focus occur excessive with flying high focus spacing, and fly high focus and cannot be carried out data analysis effectively with height change rule confusion, as shown in figure 14, in figure, curve shows, when the angle of attack is less than 4 degree, the biggest residual quantity is there is in angle of attack focus with flying high focus spacing, maximum point is more than one times of main wing chord length, as the critical piece utilizing ground effect, fly high focus mainly to be produced by main wing, but can be seen that from the focal curve of test original data processing, fly high focus beyond main wing leading edge, focus is not on main wing, this is irrational.

It is test data curve as shown in Figure 2-5, it is moment slope and lift slope ratio gained owing to flying high focus, therefore the change that each point is trickle all can cause the bigger change (becoming apparent from the slope of curve of h/b < 0.3 changes greatly strong ground effect region) of slope value, thus causes the distortion of focus data.Original experiment data c_L、m_zThere is significantly fluctuation with height change curve, particularly evident in α≤4 °, this is consistent with focus result, it was demonstrated that focus results abnormity is relevant with this.c_L、m_zPreferable with angle of attack variation rule, curve fluctuation without exception, therefore go out public transport focal curve Changing Pattern with original data processing reasonable.

Utilize above-mentioned formula that test data is carried out regression fit, obtain data after one group of matching, as shown in Fig. 6～9.C intuitively_L、m_zHeight change curve is without obvious wave phenomenon, c_L、m_zInconspicuous with angle of attack variation curvilinear motion.Wind tunnel test data are the results of objective practice, although there is error but still be the basis of data analysis, it is impossible to deviated from test data because of pursuing the smoothness of curve simply, therefore give test data and fitting data comparison curves as shown in Figure 10～13.Curve shows that fitting data curve essentially coincides with test data, c_L、m_zOriginal with fitting data correlation coefficient more than 99%, maximum data deviation about 5%, c_L、m_zStandard deviation is respectively 0.005079 and 0.003643, and after showing matching, data the most well reduce test data while improving curve law.

Matching, homing method are sketched and are: by test data c_LArranging is with height change form under different attitude angle states, as a example by lift, as shown in table 1:

Table 1 test data is with height change form

Owing to formula having three variablees a, b, c, from 5 groups of test data, therefore take three groups i.e. can determine that system number a, b, a c,Have 10 combinations ABC, ABD, ABE, ACD, ACE, ADE, BCD, BCE, BDE, CDE, then formula 1 is utilized to be calculated 10 system number a1, b1, c1.......a10, b10, c10, being combined containing 1 system number, 2 system numbers ... .10 resistance coefficient according to often organizing packet by this 10 resistance coefficient, number of combinations isOne group of mean coefficient is obtained, by obtain by packet has the group of many systems number to be averagedSystem number substitutes into formula 1 and initial data carries out regression Calculation and draws formula multiple correlation coefficient, matching, recurrence obtains each group of multiple correlation coefficient, then maximum multiple correlation coefficient is obtained, result is as shown in figure 15, Figure 15 is the analysis result that fitting data processes the focal curve figure obtained, curve shows that the result that angle of attack focus and initial data draw is more or less the same, fly the reasonability of high focal curve, dispersion and regular be improved significantly, clearly angle of attack focus and the position relationship flying high focus can be judged from stability analysis angle, judgment basis is provided for type selecting and profile refinement.

The focal curve processed out such as Figure 16, the 17 original and fitting data showing another example (profile is different), shows that the result regularity that data process after matching is obviously improved with the linearity, is more beneficial for the design of flight control rate.

The content not being described in detail in description of the invention belongs to the known technology of professional and technical personnel in the field.

Claims (1)

1. a ground effect wind tunnel test data processing method, it is characterised in that step is as follows:

(1) obtain ground effect wind tunnel test data and transform it into the data form relevant with heightWherein C_LFor lift coefficient, m_ZPitching moment coefficient,For height coefficient；

(2) test data after conversion in step (1) is carried out permutation and combination and forms multi-group data combination, often the combination of group data comprises 3 test data points；

(3) the test data point often organized in data combination substituting into formula (1) respectively and solve following formula coefficient, multi-group data combination can obtain to organize corresponding coefficient more:

$(C_L+a)\&times;(\stackrel{\&OverBar;}{h}+b)=C$

$(m_Z+a_1)\&times;(\stackrel{\&OverBar;}{h}+b_1)=C_1---\left(1\right)$

Wherein, C_LFor lift coefficient, a, b, C are lift formula generation to ask coefficient, m_ZFor pitching moment coefficient, a₁、b₁、C₁Coefficient is sought for pitching moment formula generation,For height coefficient；(4) judge whether the combination of the multi-group data in step (2) completes solving in step (3), if completing, entering step (5), otherwise entering step (3)；

(5) the many groups equation coefficients solved in step (3) is grouped, according to often organizing, data combination comprises 1 system number, 2 system numbers ..., many systems number are grouped, average coefficient value is asked in combination to including many systems number, is normalized to 1 system number；

(6) each system number solved in step (5) is substituted into formula (1) and carry out regression Calculation with original experiment data respectively, try to achieve the multiple correlation coefficient of formula；

(7) judge whether the combination of the many groups equation coefficients in step (5) completes to solve multiple correlation coefficient in step (6), if completing, entering step (8), otherwise entering step (6)；

(8) one group of multiple correlation coefficient maximum is selected as test data correction equation coefficients；

(9) utilize the test data under the correction formula correction corresponding states that step (8) obtains, revise the test data obtaining under another state after terminating and enter step (1).