JP7185565B2 - Method for estimating tire test results - Google Patents

Description

The present invention relates to a method for estimating tire test results.

As a performance evaluation test for tires used in automobiles, etc., there is a drum test in which an arbitrary road surface is attached to a drum and the tire is rolled on the drum. The drum test is also performed as an alternative to the actual vehicle test, but the test results change depending on the shape of the road surface to which the tape is attached. Therefore, it is preferable that the road surface adhered to the drum should reproduce the road surface of the actual vehicle test. However, due to problems such as the lack of information on the test course for actual vehicle tests and the cost of preparing the road surface, it is difficult to reproduce the road surface of the actual vehicle test more than a certain level on the drum test surface. In addition, when correlating performance evaluation tests conducted by third parties with performance evaluation tests conducted by oneself, road surface information from performance evaluation tests conducted by third parties may not be obtained accurately. Therefore, in such cases, the results of both tests may not match.

JP-A-2002-90264

In tire performance evaluation tests, as described above, two or more test results may differ, so there is a demand for a method of matching the test results by some means. In particular, there is a need for a method of estimating tire test results on a first, unknown road surface based on tire test results on a second, more detailed road surface.

An object of the present invention is to provide a method for estimating tire test results on a first unknown road surface by correlating a first unknown road surface with a second unknown road surface.

The present invention prepares a tire for correction examination, a test tire, a first road surface whose details are unknown, and a second road surface whose details are known, and when the tire for correction examination is run on the first road surface Acquiring a first feature amount as time-series data of, acquiring a second feature amount as time-series data when the correction examination tire is run on the second road surface, and acquiring the first feature amount and the first A first frequency analysis result and a second frequency analysis result are obtained by frequency-analyzing the two feature quantities, respectively, and are applied to the second frequency analysis result to approximate the first frequency analysis result, and the first frequency analysis result is obtained. A correction coefficient that improves the correlation between the second frequency analysis result and the second frequency analysis result is calculated at each frequency, and a test feature amount is obtained as time-series data when the test tire is run on the second road surface, By frequency-analyzing the test feature value to obtain a test frequency analysis result, and applying the correction coefficient to the test frequency analysis result at each frequency, when the test tire is run on the first road surface, A method for estimating tire test results is provided that includes estimating a frequency analysis of the resulting feature.

According to this method, it is possible to estimate the frequency analysis result of the feature quantity obtained when the test tire whose performance is to be evaluated is run on the first road surface whose details are unknown. The feature quantity may be, for example, tire axial force, moment, acceleration, or noise. This method is useful when detailed information about the first road surface is not available and it is not possible to actually test the test tire on the first road surface. As a specific method, first, a correlation between the test results of the first road surface and the test results of the second road surface is taken using a tire for correction study that is different from the test tire. Even if you cannot actually test the first road surface yourself, you may ask the owner of the first road surface to conduct the test, or you may use the past test results. Then, based on the obtained correlation, a correction coefficient is calculated so as to improve the correlation. The correction coefficient is a correction magnification for bringing the second frequency analysis result closer to the first frequency analysis result, and is set for each frequency. For example, when the spectrum of the first frequency analysis result and the spectrum of the second frequency analysis result match perfectly at a frequency of 100 Hz, the correction coefficient at a frequency of 100 Hz is 1.0. By using the correction coefficient, even if the detailed information of the first road surface is not obtained, the test result of the first road surface (test feature value test frequency analysis results) can be estimated. Here, it is preferable that the tire 1 for correction examination for acquiring a set of the first feature amount and the second feature amount is the same, but if it has the same spec even if it is not completely the same. good. It should be noted that obtaining the first frequency analysis result by frequency-analyzing the first feature quantity need not be performed by the user himself/herself, and may be performed by a third party.

In calculating the correction coefficient, the total value of the first frequency analysis result in the predetermined frequency range is calculated as a first total value, and the total value of the second frequency analysis result in the predetermined frequency range is calculated as a second total value. , the first total value and the second total value are calculated for a plurality of the correction examination tires, and the correlation coefficient is calculated from the obtained plurality of the first total value and the second total value and searching the correction factor to maximize the correlation factor.

According to this method, since a correction coefficient that maximizes the correlation coefficient is searched for, the obtained correction coefficient realizes a high correlation. Therefore, by using the correction coefficient, the frequency analysis result of the feature quantity of the first road surface can be estimated with high accuracy from the test frequency analysis result of the second road surface.

The correction coefficient may be 0.3 or more and 1.5 or less.

According to this method, the search can be simplified by limiting the search range of the correction coefficient. In practice, even if the details of the first road surface are unknown, from the overview of the first road surface (photograph of the first road surface, the purpose of the test using it, etc.), the second road surface is a road surface that simulates the first road surface to a certain extent. Assuming that it can be used, it can be inferred that the correction coefficient will not take zero, minus, or a significantly large value, so excluding the correction coefficient enables efficient searching.

In searching for the correction coefficient, the predetermined frequency range is divided into a plurality of frequency bands, the same correction coefficient is applied to the same frequency band, and the correction coefficients of adjacent frequency bands are 1/3 of each other. It may be twice or more and three times or less.

According to this method, it is possible to prevent the correction coefficients from suddenly changing in adjacent frequency bands. In practice, even if the details of the first road surface are unknown, from the overview of the first road surface (photograph of the first road surface, the purpose of the test using it, etc.), the second road surface is a road surface that simulates the first road surface to a certain extent. Given that it is possible to use it, it is difficult to imagine that the correction coefficients will change suddenly in adjacent frequency bands.

The correction coefficient is calculated by dividing the first frequency analysis result by the second frequency analysis result to obtain the correction coefficient, and obtaining the correction coefficient by the division for the plurality of correction examination tires. , averaging the correction coefficients obtained for each of the plurality of correction examination tires.

According to this method, since the correction coefficient is directly calculated by dividing the first frequency analysis result by the second frequency analysis result, a high correlation can be realized by the correction coefficient. Therefore, highly accurate estimation is possible.

As for the correction coefficient, a value of 0.9 or more and 1.1 or less may be set to 1.0.

According to this method, unnecessary correction is eliminated.

According to the present invention, since the correction coefficient is calculated by taking the correlation between the first road surface whose details are unknown and the second road surface whose details are known, the tire test result on the first road surface whose details are unknown can be estimated. .

Side view of drum testing equipment Top view of drum testing equipment Graph showing correlation before correction Graph showing corrected correlation

Embodiments of the present invention will be described below with reference to the accompanying drawings.

(First embodiment)
The tire test result estimation method of the present embodiment is a method of estimating the tire test result on the first road surface whose details are unknown by correlating the first road surface whose details are unknown and the second road surface whose details are known. . According to this method, even in a situation where the performance evaluation test of the test tire on the first road surface cannot be performed, by correcting the test result of the test tire on the second road surface, the test on the first road surface can be performed. It is possible to estimate the test results of tires for

In the tire test result estimation method of the present embodiment, a plurality of correction study tires and one test tire are used.

The plurality of correction review tires are not tires whose performance is to be tested, but tires used for calculating correction coefficients as described later. In this embodiment, 13 tires, for example, are used as tires for reviewing correction. A test tire is a tire whose performance you actually want to test. Henceforth, when there is no need to distinguish between tires for correction examination and tires for testing, they are simply referred to as tires.

FIG. 1 is a side view of a drum testing device 10 used in this embodiment. FIG. 2 is a plan view of the drum testing apparatus 10. FIG.

The drum testing device 10 is a device used to roll the tire 1 and evaluate various feature quantities. Here, the feature amount may be, for example, the axial force, moment, acceleration, or noise of the tire.

A drum testing apparatus 10 includes a drum body 11 , a shaft member 12 that supports the tire 1 , and a motor 13 that rotates the shaft member 12 . In the present embodiment, the motor 13 for rotating the shaft member 12 is provided, but the drum main body 11 is provided with the motor, and the rotation of the drum main body 11 also rotates the tire 1 that is in contact with the drum main body 11. It's okay.

The drum main body 11 is a portion where the tire 1 contacts the ground, and a road surface having a shape to be tested is attached. The road surface to be pasted is provided corresponding to the first road surface whose details are unknown or the second road surface whose details are known. Unknown details means that information such as surface unevenness cannot be obtained because the first road surface is owned by a third party. In other words, "known in detail" means that information such as surface unevenness can be accurately obtained by owning the second road surface. The drum body 11 is larger than the tire 1, and may have a diameter of 1.7m or 3.0m, for example.

The shaft member 12 is a member that serves as a rotating shaft of the tire 1 . One end of the shaft member 12 is mechanically connected to the rotation center C of the tire 1 . The other end of shaft member 12 is mechanically connected to motor 13 .

The motor 13 gives rotational power to the tire 1 via the shaft member 12 . As a result, the tire 1 rolls on the drum body 11 while being supported by the shaft member 12 .

In this embodiment, a sensor 14 is attached to the shaft member 12 . The sensor 14 measures the vertical axial force applied to the shaft member 12 . In addition, what is measured is not limited to the axial force in the vertical direction, and may be, for example, the axial force in the lateral direction or the longitudinal direction. Here, the vertical direction is the vehicle height direction when the tire 1 is mounted on the vehicle body, the horizontal direction is the vehicle width direction, and the front-rear direction is the vehicle length direction. Also, what is measured may be the moment applied to the shaft member 12 . The moment may be measured in any direction about the vertical axis, the side-to-side axis, or the front-to-back axis. Also, what is measured may be vertical acceleration, horizontal acceleration, or longitudinal acceleration, or even noise. At this time, the sensor 14 is not limited to being attached to the side surface of the shaft member 12 as shown in FIG. For example, it may be attached to the end of the shaft member 12 that is mechanically connected to the center of rotation C of the tire 1 . Moreover, when measuring noise, a sensor may be arranged at a position away from the tire 1 .

Specific steps (correlation preliminary determination step, correction coefficient calculation step, and test result estimation step) of the tire test result estimation method of the present embodiment will be described below.

(Correlation prior determination step)
The correlation prior determination step is a step of correlating the tire test results on the first road surface and the second road surface. That is, in this step, it is determined how close the first road surface and the second road surface are.

In this step, first, tire test results for 13 correction study tires are acquired using the drum test apparatuses 10 corresponding to the first road surface and the second road surface, respectively. As for the first road surface, test results conducted by a third party may be obtained, or past data may be used. As for the second road surface, you may actually test it yourself, or you may use past data.

Acquired data in the present embodiment is vertical axial force that changes in time series and is obtained when the same tire for examination of correction is run on the first road surface and the second road surface. This time-series data is obtained for the number of tires for correction review. In this embodiment, 13 pieces of data (first feature quantity) regarding the vertical axial force on the first road surface are obtained, and 13 pieces of data (second characteristic quantity) on the vertical axial force on the second road surface are obtained. That is, 13 sets of time-series data are obtained. In addition, it is preferable that the tires for correction consideration for obtaining a set of corresponding first feature amount and second feature amount are the same, but if they have the same specs even if they are not exactly the same good. The specifications refer to the tread pattern of the tire 1, size, structure of the rubber material, air pressure, load applied to the tire 1 by the drum testing device 10, and the like.

Next, frequency analysis is performed on each of the 13 sets of time-series data obtained as described above. Through frequency analysis, 13 sets of time-series data (first feature quantity and second feature quantity) are converted into 13 sets of spectra for each frequency (first frequency analysis result and second frequency analysis result). . Various known methods can be used for frequency analysis, and for example, Fourier transform may be used. Alternatively, fast Fourier transform, short-time Fourier transform, Wigner distribution, or wavelet transform may be employed, and octave analysis may be used to perform frequency analysis.

Next, a first total value Y and a second total value X are calculated as total values in a predetermined frequency range for the first frequency analysis result and the second frequency analysis result, respectively. The first total value Y and the second total value X are obtained by the following formulas.

ｆ：周波数
ｆ１：所定の周波数の下限値
ｆ２：所定の周波数の上限値
Ｓ１：第１周波数分析結果のスペクトル関数
Ｓ２：第２周波数分析結果のスペクトル関数

f: frequency f1: lower limit value of predetermined frequency f2: upper limit value of predetermined frequency S1: spectrum function of first frequency analysis result S2: spectrum function of second frequency analysis result

Alternatively, the first total value Y and the second total value X may be calculated as integral values.

The predetermined frequency range (f1≦f≦f2) may vary depending on the object to be measured, but is preferably 40-150 Hz when the vertical axial force is to be measured as in this embodiment.

Specifically, for the predetermined frequency range (f1≦f≦f2), a suitable range can be set as follows according to the object to be measured.
・40-150Hz (vertical axial force, vertical acceleration, or noise)
・180-220Hz (vertical axial force, vertical acceleration, or noise)
・0-40Hz (vertical axial force, vertical acceleration, longitudinal axial force, longitudinal acceleration, or noise)
・0-150Hz (vertical axial force, vertical acceleration, longitudinal axial force, longitudinal acceleration, or noise)
・20-50Hz (longitudinal axial force, longitudinal acceleration, lateral axial force, lateral acceleration, or noise)
・50-100Hz (moment around vertical axis or noise)
・100-170Hz (lateral axial force, lateral acceleration, moment around longitudinal axis, or noise)
・170-250Hz (vertical axial force, vertical acceleration, longitudinal axial force, longitudinal acceleration, or noise)

FIG. 3 plots 13 sets of data sets, with the first total value and the second total value obtained as described above as one set, the vertical axis being the first total value Y, and the horizontal axis The axis is the second sum X.

In FIG. 3, line segments indicating the results of linear regression are drawn together. The correlation coefficient is a value obtained by dividing the covariance of the first total value Y and the second total value X as shown in FIG. 3 by the product of the standard deviation of the first total value Y and the standard deviation of the second total value X is required as A correlation coefficient closer to 1 indicates a higher correlation, and a correlation coefficient closer to 0 indicates a lower correlation. Specifically, it is obtained by the following formula, where r is the correlation coefficient.

ｎ：相関係数の算出に使用した補正検討用タイヤの総数
Ｙ：変数（第１合計値）
Ｘ：変数（第２合計値）
ａ：データ番号（ａ=１～ｎ）

n: total number of tires for correction examination used to calculate the correlation coefficient Y: variable (first total value)
X: variable (second total value)
a: Data number (a = 1 to n)

In this manner, in the correlation preliminary determination step, the correlation between the tire test results on the first road surface and the second road surface is obtained in the form of a correlation coefficient. In the following, a correction coefficient calculation step for obtaining a correction coefficient that maximizes this correlation coefficient and a test result estimation step for estimating the test result of the test tire on the first road surface using the obtained correction coefficient will be described. .

(Correction coefficient calculation process)
In the correction coefficient calculation step, a correction coefficient that maximizes the correlation coefficient is obtained. The correction coefficient is a correction magnification for bringing the second frequency analysis result closer to the first frequency analysis result, and is set for each frequency. In this embodiment, a correlation coefficient is used to evaluate the degree of approximation.

In this step, in the predetermined frequency range (f1 ≤ f ≤ f2), various correction coefficients are integrated with respect to the second frequency analysis result, and the total value of the second frequency analysis result corrected by the integration; The correlation with the total value of the first frequency analysis result is confirmed by the correlation coefficient in the same manner as described above. Then, the correlation coefficient is checked for various correction coefficients, and the correction coefficient that maximizes the correlation coefficient is searched.

The correction coefficient to be searched may be, for example, 0.3 or more and 1.5 or less, and the step size to be changed may be 0.01. In this case, the correlation coefficient corresponding to each correction coefficient is recorded while increasing the correction coefficient by 0.01 from 0.3 to 1.5. Then, a correction coefficient that maximizes the correlation coefficient is calculated. By limiting the search range of the correction coefficient in this manner, the search can be simplified. In practice, even if the details of the first road surface are unknown, from the overview of the first road surface (photograph of the first road surface, the purpose of the test using it, etc.), the second road surface is a road surface that simulates the first road surface to a certain extent. If it can be used, it can be inferred that the correction coefficient will not take zero, minus, or a significantly large value, so excluding the correction coefficient enables efficient searching.

The above search for the correction coefficient is performed for each frequency. Specifically, a predetermined frequency range (f1≦f≦f2) may be divided into a plurality of frequency bands, and the same correction coefficient may be used in the same frequency bands. That is, if the predetermined frequency range (f1≦f≦f2) is, for example, 40-150 Hz, it is divided into 11 frequency bands every 10 Hz, and the correction coefficient is changed for each frequency band to maximize the correlation coefficient. search for a correction factor that However, the width of the frequency band is not limited to 10 Hz, and may be any width.

For example, first search for correction factors at 40-50 Hz. At this time, the correction coefficient is set to 1.0 in the range of 50-150 Hz. As a search result, for example, if it is confirmed that the correlation coefficient becomes maximum when the correction coefficient is 0.9, the correction coefficient of 40-50 Hz is set to 0.9 and the correction coefficient of 60-150 Hz is set to 1.0. and search for a 50-60 Hz correction factor. As a search result, for example, if it is confirmed that the correlation coefficient becomes maximum when the correction coefficient is 1.2, the correction coefficient of 40-50 Hz is maintained at 0.9, and the correction coefficient of 50-60 Hz is set to 1.2. , keep the 70-150 Hz correction factor at 1.0, and search for the 60-70 Hz correction factor. When such a search is completed up to 140-150 Hz, it returns to the search of 40-50 Hz and loops the search. Such a search loop is performed until the maximum value of the correlation coefficient converges, and a correction coefficient that maximizes the correlation coefficient is calculated. However, if the correlation coefficient does not converge, the search is terminated when a certain number of search loops are completed, and the correction coefficient at the maximum correlation coefficient up to that point is adopted. The upper limit of the search loop count may be 200, for example. A description such as 40-50 Hz indicates 40 or more and less than 50 or 50 or less.

In searching for the correction coefficients, the correction coefficients of adjacent frequency bands are preferably 1/3 or more and 3 or less times each other. For example, when the correction factor for 50-60 Hz is 1.2, the correction factors for 40-50 Hz and 60-70 Hz are preferably 0.4 or more and 3.6 or less. However, when the search range of the correction coefficient is 0.3 or more and 1.5 or less as described above, the correction coefficients of 40-50 Hz and 60-70 Hz are preferably 0.4 or more and 1.5 or less. Become. By defining the search for the correction coefficients in this way, it is possible to prevent the correction coefficients from suddenly changing in adjacent frequency bands. In practice, even if the details of the first road surface are unknown, from the overview of the first road surface (photograph of the first road surface, the purpose of the test using it, etc.), the second road surface is a road surface that simulates the first road surface to a certain extent. Given that it is possible to use it, it is difficult to imagine that the correction coefficients will change suddenly in adjacent frequency bands.

FIG. 4 is a graph corresponding to FIG. 3 and showing the corrected correlation. The vertical axis of the graph indicates the first total value as in FIG. The horizontal axis of the graph indicates the second total value X' calculated as the total value in a predetermined frequency range by applying the correction coefficient obtained as described above to the second frequency analysis result. Specifically, the second total value X' is obtained by the following formula.

ｆ：周波数
ｆ１：所定の周波数の下限値
ｆ２：所定の周波数の上限値
Ｓ２：第２周波数分析結果のスペクトル関数
Ｈ：補正係数

f: frequency f1: lower limit value of predetermined frequency f2: upper limit value of predetermined frequency S2: spectrum function of second frequency analysis result H: correction coefficient

Comparing FIGS. 3 and 4, in FIG. 4, since the points plotted near the line segment showing the linear regression result are concentrated, the correlation coefficient approaches 1 by correction, and the correlation is shown in FIG. It can be confirmed that it is higher than that shown.

(Test result estimation step)
In the test result estimation step, the correction coefficient calculated in the correction coefficient calculation step is used to estimate the test result (the test frequency analysis result of the test feature amount) when the test tire is run on the first road surface.

In this step, first, the vertical axial force (test feature quantity) that changes in time series when the test tire is run on the second road surface is acquired. Next, the obtained vertical axial force is frequency-analyzed, for example, by Fourier transform, and a spectrum (test frequency analysis result) at each frequency is obtained. Then, by multiplying the test frequency analysis result by the correction coefficient calculated in the correction coefficient calculation step, the frequency analysis result of the vertical axial force obtained when the test tire is run on the first road surface is estimated. .

(Effect)
According to the tire test result estimation method of the present embodiment, the following effects are obtained.

It is possible to estimate the frequency analysis result of the feature quantity obtained when the test tire whose performance is to be evaluated is run on the first road surface whose details are unknown. The method of this embodiment is useful when detailed information about the first road surface is not available and it is not possible to actually test the test tire on the first road surface. In this embodiment, by using the correction coefficient, the test result of the first road surface can be obtained from the test result of the second road surface (test feature amount and test frequency analysis result) even if detailed information of the first road surface cannot be obtained. can be estimated.

Also, since the correction coefficient that maximizes the correlation coefficient is searched for, the obtained correction coefficient realizes a high correlation. Therefore, by using the correction coefficient, the frequency analysis result of the feature quantity of the first road surface can be estimated with high accuracy from the test frequency analysis result of the second road surface.

(Second embodiment)
The tire test result estimation method of the present embodiment differs from that of the first embodiment in the correction coefficient calculation method. The method for estimating the tire test result of the first embodiment is substantially the same except for the part related to this. Therefore, explanations overlapping with the first embodiment may be omitted.

Since the correlation pre-determination step and the test result estimation step of this embodiment are the same as those of the first embodiment, description thereof will be omitted.

In the correction coefficient calculation step of the present embodiment, the correction coefficient is obtained by dividing the first frequency analysis result by the second frequency analysis result in a predetermined frequency range. The predetermined frequency range is the same as in the first embodiment. Therefore, when measuring the vertical axial force as in the first embodiment, the predetermined frequency range may be 40-150 Hz. In this case, the first frequency analysis result and the second frequency analysis result are the results of frequency analysis of the vertical axial force time series data as in the first embodiment, and are spectra at each frequency.

In this step, the correction coefficients are obtained by the above division for a plurality of tires for correction consideration, and the correction coefficients obtained for each of the plurality of tires for correction examination are averaged. For example, the correction coefficient H1(f) is obtained for the first tire for correction examination, the correction coefficient H2(f) is obtained for the second tire for correction examination, and in this way 13 correction examination tires Once the correction factors H1(f)-H13(f) are obtained for the tire, in this step these correction factors H1(f)-H13(f) are averaged. Then, the averaged correction coefficient Ht(f) is adopted as a correction coefficient to be applied to the test tire.

For the averaged correction coefficient Ht(f), a value of 0.9 or more and 1.1 or less may be set to 1.0. Also, for each of the correction coefficients H1(f) to H13(f), a value of 0.9 or more and 1.1 or less may be set to 1.0. This eliminates unnecessary correction.

According to this embodiment, since the correction coefficient is directly calculated by dividing the first frequency analysis result by the second frequency analysis result, a high correlation can be realized by the correction coefficient. Therefore, highly accurate estimation is possible.

As described above, specific embodiments of the present invention have been described, but the present invention is not limited to the above embodiments, and various modifications can be made within the scope of the present invention.

For example, in the above embodiment, the tire test is a drum test, but the tire test may be an actual vehicle test. That is, either or both of the first road surface and the second road surface may be the actual road surface instead of the road surface attached to the drum. It should be noted that obtaining the first frequency analysis result by frequency-analyzing the first feature quantity need not be performed by the user himself/herself, and may be performed by a third party.

Further, the method of the above embodiment may be automatically analyzed on a computer, or may be calculated manually.

1 tire (test tire, tire for correction review)
REFERENCE SIGNS LIST 10 drum testing device 11 drum body 12 shaft member 13 motor 14 sensor

Claims (6)

Prepare a correction study tire, a test tire, a first road surface whose details are unknown, and a second road surface whose details are known,
Acquiring a first feature value as time-series data when the correction examination tire is run on the first road surface,
Acquiring a second feature amount as time-series data when the correction examination tire is run on the second road surface,
Obtaining a first frequency analysis result and a second frequency analysis result by frequency-analyzing the first feature amount and the second feature amount, respectively;
calculating a correction coefficient at each frequency that is applied to the second frequency analysis result to approximate the first frequency analysis result and improve the correlation between the first frequency analysis result and the second frequency analysis result;
Acquiring a test feature value as time-series data when the test tire is run on the second road surface,
obtain a test frequency analysis result by frequency-analyzing the test feature quantity;
estimating the frequency analysis result of the feature quantity obtained when the test tire is run on the first road surface by applying the correction coefficient to the test frequency analysis result at each frequency. Method of estimating test results. The calculation of the correction coefficient is
calculating a total value in a predetermined frequency range of the first frequency analysis result as a first total value;
calculating a total value in the predetermined frequency range of the second frequency analysis result as a second total value;
calculating the first total value and the second total value for a plurality of the correction examination tires;
calculating a correlation coefficient from the plurality of first total values and the second total values obtained;
The method of estimating tire test results of claim 1, comprising: searching the correction factor to maximize the correlation factor. The method for estimating tire test results according to claim 2, wherein the correction factor is greater than or equal to 0.3 and less than or equal to 1.5. In searching for the correction coefficient,
dividing the predetermined frequency range into a plurality of frequency bands;
applying the same correction factor in the same frequency band;
The tire test result estimation method according to claim 3, wherein the correction coefficients of the adjacent frequency bands are 1/3 or more and 3 or less times each other. The calculation of the correction coefficient is
dividing the first frequency analysis result by the second frequency analysis result to obtain the correction factor;
Acquisition of the correction coefficient by the division in the plurality of tires for correction examination,
The method for estimating tire test results according to claim 1, comprising averaging the correction coefficients obtained for each of the plurality of tires for correction examination. 6. The tire test result estimation method according to claim 5, wherein the correction coefficient is 1.0 when the value is 0.9 or more and 1.1 or less.

Images