CN103473470A - Ground effect wind tunnel test data processing method - Google Patents

Abstract

A ground effect wind tunnel test data processing method, the present invention uses the data correction formula applicable to the ground effect and the coefficient regression solution calculation method to quickly and objectively correct the wind tunnel test data with interference; the method will test Bring the data into the correction formula to find the coefficients in the formula, use the regression algorithm to find a set of coefficients with the highest correlation with the original data as the coefficients of the correction formula, and then use the determined formula to calculate the data that is the same as the state of the test point. The results of this method show that the corrected data has a high degree of restoration, and can provide regular basic data for post-processing and analysis of data, and create good conditions for the shape selection and fine design of ground-effect aircraft.

Description

A data processing method for ground effect wind tunnel test

technical field

The invention relates to a post-processing method for ground-effect wind tunnel test data, which is used for analyzing the near-ground stability of ground-effect aircraft and belongs to the technical field of aircraft testing and testing.

Background technique

The general ground effect test is to study the aerodynamic characteristics of the aircraft at different heights from the ground, so its aerodynamic data variable parameter has an additional height variable h compared with the conventional aircraft. Declination test, the test data includes in addition to the conventional c _L ～α, m _z ～α The above data are used to calculate the "angle of attack focus" and "fly height focus". By 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-geostatic stability conditions:

的精度决定了两个焦点数据的计算精度。常规试验(不含地面效应)技术成熟、干扰因素小且易于消除，一般能够得到精度较高的平顺曲线；而加入地面边界条件后，由于试验技术的限制无法真实模拟地板与飞行的的相对运动及边界层情况，因此试验中干扰较为严重，得到的

曲线往往存在奇异点且曲线平顺度不好（失真），会严重影响

求解精度，造成“飞高焦点”曲线波动较大无法准确评估近地稳定性。It can be seen that the experimental data c _L ～α, m _z ～α,

The precision of determines the calculation precision of the two focus data. Routine tests (excluding ground effects) have mature technology, small interference factors and are easy to eliminate. Generally, smooth curves with high precision can be obtained; however, after adding ground boundary conditions, due to the limitation of test technology, the relative motion between the floor and the flight cannot be truly simulated. and boundary layer conditions, so the interference is more serious in the test, and the obtained

There are often singular points in the curve and the smoothness of the curve is not good (distortion), which will seriously affect

The accuracy of the solution results in large fluctuations in the "flying high focus" curve, which cannot accurately evaluate the near-Earth stability.

Contents of the invention

The technical problem of the present invention is: to overcome the deficiencies of the prior art, provide a ground effect wind tunnel test data processing method, eliminate the test interference of test data analysis, and provide conditions for accurate analysis of stability.

Technical solution of the present invention is:

A ground effect wind tunnel test data processing method, comprising the following steps:

(1) Obtain ground effect wind tunnel test data and transform it into a data format related to height

(2) Arrange and combine the transformed test data in step (1) to form multiple sets of data combinations, each of which contains 3 test data points;

(3) Substituting the test data points in each group of data combinations into formula (1) to solve the coefficients of the following formula, multiple groups of data combinations can obtain multiple groups of corresponding coefficients:

$<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; C</span>}$

$<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

为高度系数；（4）判断步骤（2）中的多组数据组合是否完成步骤（3）中的求解，若完成则进入步骤（5），否则进入步骤（3）；Among them, C _y is the lift coefficient, a, b, C are the substitution coefficients of the lift formula, m _Z is the pitching moment coefficient, a ₁ , b ₁ , C ₁ are the pitching moment formula substitution coefficients,

is the height coefficient; (4) judge whether the combination of multiple sets of data in step (2) has completed the solution in step (3), if it is completed, go to step (5), otherwise go to step (3);

(5) Group the multiple sets of formula coefficients solved in step (3), and group them according to each set of data containing 1 set of coefficients, 2 sets of coefficients, ... multiple sets of coefficients, and group the sets containing multiple sets of coefficients Find the average coefficient value and normalize it into a set of coefficients;

(6) Substitute each group of coefficients solved in step (5) into formula (1) and perform regression calculation with the original test data respectively to obtain the multiple correlation coefficient of the formula;

(7) Judging whether the combination of multiple formula coefficients in step (5) has completed the solution of the complex correlation coefficient in step (6), if it is completed, go to step (8), otherwise go to step (6);

(8) Select the group with the largest multiple correlation coefficient as the formula coefficient for the correction of test data;

(9) Use the correction formula obtained in step (8) to correct the test data in the corresponding state. After the correction is completed, obtain the test data in another state and enter step (1).

The beneficial effect of the present invention compared with prior art is:

(1) Compared with the original technology, the present invention does not correct the data or only fine-tune individual singular points (other data are not adjusted), which ensures the systematicness of the adjusted data and improves the stability of the entire data processing.

(2) Compared with the original technology, the present invention relies on manual adjustment. The present invention forms a complete set of data correction methods and procedures, which can achieve objective and fast completion of data correction.

(3) The data correction method of the present invention is simple and intuitive, and is more suitable for ground effect data processing than high-order interference correction techniques such as Fourier transform.

Description of drawings

Fig. 1 is a flow chart of the method of the present invention;

Fig. 2 is the test curve of lift force varying with height;

Fig. 3 is the test curve of moment varying with height;

Fig. 4 is the test curve of the lift force varying with the angle of attack;

Fig. 5 is the experimental curve of torque changing with angle of attack;

Fig. 6 is the fitting curve of lift changing with height;

Fig. 7 is the fitting curve of torque changing with height;

Fig. 8 is the fitting curve of lift changing with angle of attack;

Fig. 9 is the fitting curve of torque changing with angle of attack;

Figure 10 is a comparison curve of lift force with height;

Figure 11 is a comparison curve of torque with height;

Fig. 12 is the comparison curve of lift force changing with angle of attack;

Fig. 13 is a comparison curve of torque with angle of attack;

Figure 14 is the original focus map;

Figure 15 is a fitted focus map;

Figure 16 is the original focus map;

Figure 17 is a fitted focal map.

Detailed ways

Specific embodiments of the present invention will be further described in detail below in conjunction with the accompanying drawings.

As shown in Fig. 1, a kind of ground effect wind tunnel test data processing method, the present invention is that the data that contains disturbance that test obtains is carried out fitting process, obtains the curve that the rule that changes smoothly based on test data, is follow-up stable Provide favorable conditions for performance analysis and control rate design. The present invention comprises steps as follows:

其中m_Z为俯仰力矩系数，

为高度系数；(1) Obtain ground effect wind tunnel test data and transform it into a data format related to height

where m _Z is the pitching moment coefficient,

is the height coefficient;

(2) Arrange and combine the transformed test data in step (1) to form multiple sets of data combinations, each of which contains 3 test data points;

(3) Substituting the test data points in each group of data combinations into formula (1) to solve the coefficients of the following formula, multiple groups of data combinations can obtain multiple groups of corresponding coefficients:

$<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; C</span>}$

$<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

为高度系数；Among them, C _L is the lift coefficient, a, b, C are the substitution coefficients of the lift formula, m _Z is the pitching moment coefficient, a ₁ , b ₁ , C ₁ are the pitching moment formula substitution coefficients,

is the height coefficient;

(4) Judging whether the combination of multiple sets of data in step (2) has completed the solution in step (3), if it is completed, go to step (5), otherwise go to step (3);

(5) Group the multiple sets of formula coefficients solved in step (3), and group them according to each set of data containing 1 set of coefficients, 2 sets of coefficients, ... multiple sets of coefficients, and group the sets containing multiple sets of coefficients Find the average coefficient value and normalize it into a set of coefficients;

(6) Substitute each group of coefficients solved in step (5) into formula (1) and perform regression calculation with the original test data respectively to obtain the multiple correlation coefficient of the formula;

(7) Judging whether the combination of multiple formula coefficients in step (5) has completed the solution of the complex correlation coefficient in step (6), if it is completed, go to step (8), otherwise go to step (6);

(8) Select the group with the largest multiple correlation coefficient as the formula coefficient for the correction of test data;

(9) Use the correction formula obtained in step (8) to correct the test data in the corresponding state. After the correction is completed, obtain the test data in another state and enter step (1).

In theory, three sets of data can be used to calculate the coefficients (a, b, c) in a set of fitting formulas. Therefore, the coefficients in multiple sets of formulas can be calculated by using the original test data. All the calculated coefficients, including the single set of coefficients, can be calculated. And the average value of several sets of coefficients is brought into the formula for regression calculation, and the set of coefficients with the highest multiple correlation coefficient with the original data is taken as the coefficient of the fitting formula, which is used to calculate the corrected data.

The working process of the present invention is further described below with a specific example. According to the test results of a ground-effect aircraft, the focal relationship diagram obtained from the experimental data processing shows that the distance between the focus of the angle of attack and the focus of the flying height is too large in the range of small angles of attack (α≤4°), and the focus of the flying height increases with the altitude. The changing law is chaotic and cannot be used for effective data analysis. As shown in Figure 14, the curve in the figure shows that when the angle of attack is less than 4 degrees, there is a large difference in the distance between the focal point of the attack angle and the focal point of the flying height, and the maximum point has exceeded one time of the chord length of the main wing , as the main component using ground effect, the flying height focus is mainly generated by the main wing, but from the focus curve processed by the test raw data, it can be seen that the flying height focus has exceeded the leading edge of the main wing, and the focus is not on the main wing, which is unreasonable of.

The test data curves are shown in Figures 2 to 5. Since the focus of the flying height is obtained by the ratio of the slope of the moment to the slope of the lift, a slight change in each point will cause a large change in the slope value (the slope of the curve with h/b<0.3 changes It is more obvious in the larger and stronger ground effect area), which causes the distortion of the focal point data. There are obvious fluctuations in the curves of the original test data c _L , m _z with height, especially within α≤4°, which is consistent with the focus processing results, proving that the abnormal focus results are related to this. c _L , m _z change well with the angle of attack, and the curves have no abnormal fluctuations, so the change law of the bus focus curve obtained by processing the original data is reasonable.

Use the above formula to perform regression fitting on the experimental data, and obtain a set of fitted data, as shown in Figures 6-9. Intuitively, the c _L , m _z height change curves have no obvious fluctuations, and the c _L , m _z change curves with the angle of attack do not change significantly. The wind tunnel test data is the result of objective practice. Although there are errors, it is still the basis of data analysis. It cannot deviate from the test data because of blindly pursuing the smoothness of the curve. Therefore, the test data and simulated Combined data comparison curve. The curve shows that the fitted data curve basically coincides with the test data, the correlation coefficient between the original and fitted data of c _L , m _z is above 99%, the maximum data deviation is about 5%, and the standard deviations of c _L , m _z are 0.005079 and 0.003643, respectively. It shows that the fitted data can restore the experimental data well while improving the curve law.

The fitting and regression methods are briefly described as follows: the test data c _L is sorted into forms that vary with height under different attitude angles, taking the lift as an example, as shown in Table 1:

Table 1 The variation form of test data with height

共有10个组合ABC、ABD、ABE、ACD、ACE、ADE、BCD、BCE、BDE、CDE，然后利用公式1计算得到10组系数a1、b1、c1.......a10、b10、c10，将这10阻系数按照每组数据包含1组系数、2组系数、....10阻系数进行组合，组合个数为

将分组中有多组系数的组进行平均得到一组平均系数，将得到的组系数代入公式1与原始数据进行回归计算得出公式复相关系数，拟合、回归得到各组复相关系数，然后得到最大复相关系数，处理结果如图15所示，图15为拟合数据处理得到的焦点曲线图的分析结果，曲线表明攻角焦点与原始数据得出的结果相差不大，飞高焦点曲线的合理性、离散度及规律性得到明显改善，从稳定性分析角度可以很明显的判定攻角焦点与飞高焦点的位置关系，为选型及外形细化设计提供判定依据。Since there are three variables a, b, and c in the formula, a set of coefficients a, b, and c can be determined by taking three sets from the five sets of test data,

There are 10 combinations ABC, ABD, ABE, ACD, ACE, ADE, BCD, BCE, BDE, CDE, and then use formula 1 to calculate 10 sets of coefficients a1, b1, c1...a10, b10, c10 , the 10 resistance coefficients are combined according to each set of data including 1 set of coefficients, 2 sets of coefficients, ... 10 resistance coefficients, and the number of combinations is

Average the groups with multiple sets of coefficients in the group to obtain a set of average coefficients, and the obtained Substituting the group coefficients into formula 1 and performing regression calculation with the original data to obtain the complex correlation coefficient of the formula, fitting and regression to obtain the complex correlation coefficient of each group, and then obtaining the maximum complex correlation coefficient, the processing results are shown in Figure 15, and Figure 15 is the fitted data The analysis results of the processed focus curves show that the focus of the angle of attack is not much different from the results obtained from the original data, and the rationality, dispersion and regularity of the focus curve of the flying height have been significantly improved. From the perspective of stability analysis, it can be greatly improved. Obvious determination of the positional relationship between the focus of the angle of attack and the focus of the fly height provides a basis for determination for model selection and shape refinement design.

Figures 16 and 17 show the focal curves processed from the original and fitted data of another example (with different shapes), indicating that the regularity and linearity of the data processing results after fitting have been significantly improved, which is more conducive to Design of flight control rate.

The content that is not described in detail in the specification of the present invention belongs to the well-known technology of those skilled in the art.

Claims (1)

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

(1) obtain ground effect wind tunnel test data and it is transformed to and highly relevant data layout

c wherein _lfor lift coefficient, m _zpitching moment coefficient, for height coefficient;

(2) test figure after conversion in step (1) is carried out to permutation and combination and form the multi-group data combination, in every group of data combination, comprise 3 test figure points;

(3) the test figure point difference substitution formula (1) in every group of data combination is solved to the following formula coefficient, the multi-group data combination can be obtained the corresponding coefficients of many groups:

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

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

Wherein, C _lfor lift coefficient, a, b, C are that the lift formula is for asking coefficient, m _zfor pitching moment coefficient, a ₁, b ₁, C ₁for in pitching moment formula generation, asked coefficient, for height coefficient; (4) multi-group data in determining step (2) combines solving in completing steps (3) whether, enters step (5) if complete, otherwise enters step (3);

(5) the many groups formula coefficient solved in step (3) is divided into groups, according to the combination of every group of data comprise 1 group of coefficient, 2 groups of coefficients .... organize coefficient more and divided into groups, the combination that includes many group coefficients is asked for to average numerical value, be normalized to 1 group of coefficient;

(6) the coefficient substitution formula (1) of respectively organizing solved in step (5) is returned to calculating with original experiment data respectively, try to achieve the multiple correlation coefficient of formula;

(7) the many groups formula coefficient in determining step (5) combines the multiple correlation coefficient that solves in completing steps (6) whether, enters step (8) if complete, otherwise enters step (6);

(8) select a group of multiple correlation coefficient maximum as test figure correction formula coefficient;

(9) test figure under the correction formula correction corresponding states of utilizing step (8) to obtain, revise after finishing the test figure of obtaining under another state and enter step (1).

Images