Simple method for evaluating building wind noise

Abstract

To provide an evaluation method that can simply and accurately evaluate wind noise.SOLUTION: A simple method for evaluating building wind noise includes: determining a wind noise index having a simple pure tone degree and the total value of the simple pure tone degree and a noise level as reference values; measuring a sound pressure level value and the noise level for every 1 / n octave band frequency (n is a natural number of 1-24) when a test piece is present and when a test piece is absent in a wind tunnel experiment room, and when the difference in sound pressure level value between the presence and absence of the test piece (presence minus absence) exceeds a prescribed value, determining a simple pure tone degree (PR') or a corrected simple pure tone degree (PR'r) from the sound pressure level value with the presence of the test piece; and ranking the total value of the maximum value of the simple pure tone degree or the corrected simple pure tone degree (PR'M) and the noise level with the presence of the test piece (PR'M+LA) on the basis of the wind noise index.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to a simple method for evaluating wind noise generated by a building.

Background Art

[0002] A pure tone is a sound consisting of a single sine wave. Wind noise has a characteristic that the sound pressure level is more prominent in a narrow frequency band similar to a pure tone than in other bands, and it gives a stronger annoying impression compared to broadband noise, so its reduction is strongly demanded. In the field of architecture, in the design drawings of many buildings, it is required to examine whether wind noise is generated and whether the generated wind noise is likely to be a problem as an audible impression. However, the actual situation is that there is no established method proposed by academic societies or the like for measuring and evaluating wind noise related to buildings. Therefore, the evaluation of wind noise is carried out by a method of letting the donor of the building or the like listen to the sound when wind is applied to the actual exterior material or a mock-up model (model) cut out a part of it in a wind tunnel laboratory at various wind directions and wind speeds, and making a judgment on the spot as good or bad. This method requires a lot of time to examine the conditions of wind direction and wind speed to be measured, and it is a sensory evaluation by donors or the like who have no or little evaluation experience, and it is difficult to say that the reliability of the evaluation results is high. In addition, since residents who actually use the building usually have no opportunity to participate in this evaluation, wind noise may become a problem after actual residence.

[0003] The present inventors have proposed an evaluation method for objectively evaluating wind noise in Patent Document 1. By the method described in Patent Document 1, it is possible to objectively evaluate pure-tone wind noise, and it is possible to change the design to reduce wind noise. Here, as a method for selecting high-purity noise from actual noise containing sounds of various frequencies, the method of ISO / PAS 20065:2016 (Non-Patent Document 1) has been established. This method selects high-purity sounds using tonal audibility (TA), and the sound is more prominent than the sounds in adjacent frequency bands with similar sound pressure levels. That is, in the frequency-sound pressure level spectrum, a certain slope or more is required.

[0004] Actual wind noise also includes noise with low tonal audibility, that is, noise in a slightly wider band than pure tones. Such noise has a mountain-shaped peak in the frequency-sound pressure level spectrum, and its slope is insufficient, so it may not be selected by the method described in Non-Patent Document 1. And in the method of Patent Document 1, such noise with a mountain-shaped peak is considered "not requiring consideration because the convex slope is below the specified value (see Figure 1 of Patent Document 1)" in Step C of its evaluation flow, and the wind noise could not be evaluated.

[0005] As one method for detecting prominent discrete frequency sounds, the prominence ratio (PR) is known (Non-Patent Document 2). Although this method can detect low-purity mountain-shaped peaks, this method analyzes the spectrum of wind noise and requires a high-performance PC equipped with a specialized program or the like. In addition, since the method described in Patent Document 1 also requires processing of spectrum data to select high-purity noise, it was not possible to simply evaluate wind noise.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Non-Patent Documents

[0007]

Non-Patent Document 1

[0008] The inventors have found a method for evaluating wind noise by using a simple pure - tone audibility (PR'), which is an index representing how much the sound - pressure level at the center frequency of a certain 1 / n - octave band protrudes from the sound - pressure levels of the frequency bands adjacent above and below. With this, even a peak in the shape of a mountain that cannot be selected by the pure - tone audibility (TA) can be selected. However, upon further study, it was discovered that the simple pure - tone audibility (PR') significantly deviates in value when the peak frequency of the wind noise occurs near the center frequency of the 1 / n - octave band and when it occurs near the mid - point between the center frequencies of the two - band 1 / n - octave bands. This is because the wind noise is calculated by being divided into the center frequencies of the two - band 1 / n - octave bands.

[0009] As a result of intensive research, the inventors have found a method to correct this and have completed the present invention. That is, an object of the present invention is to provide an evaluation method that can simply and accurately evaluate wind noise. Another object of the present invention is to provide an evaluation method that can evaluate wind noise having a bandwidth wider than that of a pure tone. [Means for Solving the Problems]

[0010] The means for solving the problems of the present invention are as follows. 1. Define a wind noise index with the simple pure tone and the total value of the simple pure tone and the noise level as reference values. Measure the sound pressure level values and the noise level for each 1 / n octave band frequency (n is a natural number from 1 to 24) with and without a test body in a wind tunnel laboratory. When the difference (with - without) in the sound pressure level values with and without a test body exceeds the specified value, For all the sound pressure level values with a test body, calculate the arithmetic mean value (L k ) of the center frequencies (f k-1 , f k+1 ) above and below the center frequency (f ave_k = (L k+1 + L k-1 ) / 2), Subtract the sound pressure level value (L ave_k ) at the center frequency (f k ) from the arithmetic mean value (L k ) to obtain a numerical value (ΔL k = L ave_k - L k ). Round this numerical value to the nearest integer at the first decimal place. (1) If there is no place where the consecutive numerical values are arranged in the order of positive, negative, negative, positive (where one of the negative values may be 0), calculate the simple pure tone (PR’) based on the following formula 1 from the sound pressure level values with a test body.

Equation

[0011] 2. Defining a wind noise index with the simple pure tone and the sum of the simple pure tone and the noise level as reference values, Measuring the sound pressure level values and the noise level for each 1 / n octave band frequency (n is a natural number from 1 to 24) related to the wind noise of an actual building, For all the sound pressure level values, the center frequencies (f k ) above and below the center frequency (fk-1 , f k+1 ) arithmetic mean value (L ave_k =(L k+1 +L k-1 ) / 2) is calculated, and from the arithmetic mean value (L ave_k ), the sound pressure level value (L k ) at the center frequency (f k ) is subtracted, and the resulting value (ΔL k =L ave_k -L k ) is rounded to the nearest integer at the first decimal place, and (1) When there is no place where the consecutive values (ΔL k ) are arranged in the order of positive, negative, negative, positive (where one of the negative values may be 0), the simple pure tone degree (PR’) of the wind noise is calculated from the sound pressure level value based on the following formula 1,

Equation

Equation

Number

[0012] 3. The simple evaluation method for building wind noise according to 1. or 2., characterized in that n is any one of 3, 6, 9, and 12. 4. A building design method, characterized by feeding back the evaluation obtained by the simple evaluation method for building wind noise according to any one of 1. to 3., repeating the design of the building, and designing a building with reduced wind noise.

Effect of the Invention

[0013] The method of the present invention can evaluate wind noise from the simple phonetic degree or corrected simple phonetic degree calculated from the sound pressure level values for each 1 / n octave band frequency and the noise level. The method of the present invention can evaluate wind noise only with a general noise meter and a PC equipped with general spreadsheet software, and no special device or program is required. The method of the present invention can also evaluate wind noise without conducting a wind tunnel test.

Brief Description of Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Mode for Carrying Out the Invention

[0015] · Simple sound pressure level (PR’) The simple sound pressure level (PR’) is a value calculated by the following formula 1 based on the sound pressure level values at the 1 / n octave band center frequencies (n is a natural number from 1 to 24).

Number

[0016] In the present invention, n is a natural number from 1 to 24. The value of n is not particularly limited, but 3, 6, 9, 12 are preferred, 3, 6 are more preferred, and 3 is most preferred. The 1 / 3 octave is defined in JIS C1513-1:2020, and the sound pressure level at the 1 / 3 octave band center frequency can be measured with a general sound level meter.

[0017] As shown in formula 1, the simple sound pressure level (PR’) is the sound pressure level (L M ) at a certain 1 / n octave band center frequency (f M ’), the center frequency (f L ) of the adjacent lower frequency band, and the center frequency (fU ) sound pressure level (each of L L ’, L U ’) is the difference from the average, and is an index indicating how much it protrudes from the adjacent frequency bands. With this simple pure tone degree (PR’), it is possible to select a peak in the shape of a mountain that cannot be selected by the pure tone audibility (TA). Note that the simple pure tone degree (PR’) can also select a peak in a narrow band that can be detected by the pure tone audibility (TA).

[0018] · Corrected simple pure tone degree (PR’r) When the peak frequency of the wind noise occurs near the center frequency of the 1 / n octave band, and when it occurs near the middle of the center frequencies of the 1 / n octave bands of two bands, there is a significant deviation in the value of the simple pure tone degree (PR’). This is because when the peak frequency of the wind noise is located near the middle of the center frequencies of the 1 / n octave bands of two adjacent bands, it is divided into the center frequencies of these two adjacent 1 / n octave bands. In principle, this deviation is more likely to occur as the frequency band is wider, that is, as the value of n is smaller. Therefore, the evaluation method of the present invention is more suitable when the value of n is small. n is preferably 12 or less, more preferably 6 or less, further preferably 5 or less, even more preferably 4 or less, and even more preferably 3 or less.

[0019] For example, when trying to evaluate the wind noise with a noise level of 78 dB and almost the same auditory impression as shown in FIG. 1 and Table 1 by the analysis value based on the center frequency of the 1 / 3 octave band, the wind noise with a peak frequency of 400 Hz has a simple pure tone degree (PR’) calculated as 15.4 dB, and the evaluation rank of the wind noise described later is “4”. On the other hand, the wind noise with a peak frequency of 447 Hz, which is near the middle of the two 1 / 3 octave band center frequencies (400 Hz and 500 Hz), is divided into the two 1 / 3 octave band center frequencies of 400 Hz and 500 Hz. Therefore, the simple pure tone degree (PR’) is calculated as 3.2 dB, and the evaluation rank of the wind noise remains “2” as it is.

Table 1

[0020] For nine cases where the peak frequency of the wind noise is located between 800 Hz and 1000 Hz, which is the center frequency of the 1 / 3 octave band, the results of obtaining the maximum value of the simple pure tone degree (PR’) are shown in Fig. 2. As the peak frequency moves away from the center frequency of the 1 / 3 octave band (1→5, 9→5 in Fig. 2), actually one peak is divided into two peaks, and the maximum value of PR’ decreases from about 14 to about 4.

[0021] Fig. 3 shows a schematic diagram of the 1 / n octave band analysis value when the peak frequency is located near the middle of the center frequencies of two 1 / n octave bands and the arithmetic mean value of the sound pressure levels at the center frequencies above and below it. As shown in Fig. 3, when the peak frequency is located near the middle of the center frequencies of two 1 / n octave bands, the values of the continuous sound pressure levels are in the order of low, high, high, low. For the nine cases in Fig. 2, regarding the sound pressure level values at the center frequency (f k ), the arithmetic mean value (L ave_k =(L k+1 +L k-1 ) / 2) of the sound pressure levels at the center frequencies (f k-1 , f k+1 ) above and below it is calculated, and the numerical value (ΔL k =L ave_k -L k ) obtained by subtracting the sound pressure level value (L k ) at the center frequency (f k ) from the arithmetic mean value (L ave_k ) is obtained as an integer rounded to the first decimal place, and the result is as shown in Table 2.

Table 2

[0022] When the values of the sound pressure level are continuous in the order of low, high, high, low, this numerical value (ΔL kThey are consecutive in the order of positive, negative, negative, and positive. Examples 4 to 8 fall under this relationship. Here, as shown in FIG. 2, since the decrease amounts of the maximum value of PR’ in Examples 3 and 8 are equal, even when one of the negative values is 0 among positive, negative, negative, and positive, it is necessary to include it in the procedure for obtaining the corrected simple phoneticness (PR’r).

[0023] Then, by utilizing this characteristic, the peaks divided into two 1 / n octave bands can be identified. (1) When there is no place where consecutive numerical values (ΔL k ) are arranged in the order of positive, negative, negative, and positive (where one of the negative values may be 0), the simple phoneticness (PR’) calculated based on the above formula 1 is used. This is because there are no peaks divided into two bands.

[0024] (2) When there is a place where consecutive numerical values (ΔL k ) are arranged in the order of positive, negative, negative, and positive (where one of the negative values may be 0), the corrected simple phoneticness (PR’r) calculated based on the following formula 2 from the sound pressure level values with the test object in these four frequency bands is used.

Equation

[0025]

Number

[0026] The corrected simple phonetic degree (PR’r) is the sound pressure level (L L , f U ) synthesized from the sound pressure levels divided into the two middle bands among the four sound pressure levels, and the difference from the average of the sound pressure levels (L MS ’), the center frequency (f L-1 ) of the low-frequency band located above and below it, and the center frequency (f U+1 ) of the high-frequency band (L L-1 ’, L U+1 ’) respectively. The corrected simple phonetic degree (PR’r) is the simple phonetic degree corrected by synthesizing two sound pressure level values that have been divided into two and become smaller into one large sound pressure level value (corrected simple phonetic degree). From the wind noise where the simple phonetic degree (PR’) is calculated as a value smaller than the actual sound pressure level, the corrected simple phonetic degree (PR’r) that more accurately reflects the actual sound pressure level can be calculated.

[0027] · Wind noise index Table 3 shows an example of the wind noise index. This wind noise index uses Rank 3 as the specified value, and it is determined that wind noise that may be a problem is occurring when the rank is 3 or higher.

Table 3

[0028] The wind noise index of the present invention is the value of the simple phonetic degree or the corrected simple phonetic degree (PR’orPR’r), and the total value (PR’orPR’r + L A ) of the simple phonetic degree or the corrected simple phonetic degree (PR’orPR’r) and the noise level (L A) is created from both of them. The wind noise index, which is an example shown in Table 3, models wind noise with mountain-shaped peaks and artificially creates a total of 52 test sounds by changing the center frequency, peak spread width, etc. Fifteen researchers in the field of building noise listened to these sounds and evaluated their annoyance on a 6-point scale from 0 to 5 as shown below. From the evaluation results, the simple pure tone loudness (PR’) or the corrected simple pure tone loudness (PR’r), and the sum of the simple pure tone loudness or the corrected simple pure tone loudness (PR’orPR’r) and the noise level (L A : A-Weighted Sound Pressure Level) ((PR’orPR’r)+L A ) has a defined threshold value. Note that the reference value and ranking of this wind noise index are only examples, and the present invention is not limited thereto.

[0029] (Evaluation Rank) 0: No sound 1: Slight sound that can be confirmed by listening carefully 2: Faintly audible sound 3: Clearly audible sound 4: Clear and highly likely to be obstructive sound 5: Very noisy and clearly obstructive sound

[0030] The present invention is a method for simply evaluating wind noise in a building. · First simple evaluation method The first simple evaluation method of the present invention is defining a wind noise index with the simple pure tone loudness and the sum of the simple pure tone loudness and the noise level as reference values, measuring the sound pressure level values and the noise level for each 1 / n octave band frequency (n is a natural number from 1 to 24) with and without a test body in a wind tunnel test chamber. When the difference (with - without) between the sound pressure level values with and without a test body exceeds a specified value, for all the sound pressure level values with a test body, the arithmetic mean value (L k ) of the center frequencies (f k-1 , f k+1 ) above and below the center frequency (f ave_k =(L k+1 +Lk-1 ) / 2) is calculated, from the arithmetic mean value (L ave_k ) the sound pressure level value (L k ) at the center frequency (f k ) is subtracted to obtain a numerical value (ΔL k = L ave_k - L k ), which is rounded to the nearest integer at the first decimal place,

[0031] (1) When there is no place where the consecutive numerical values are arranged in the order of positive, negative, negative, positive (where one of the negative values may be 0), the simple pure tone degree (PR’) is obtained from the sound pressure level value with the test specimen based on the following formula 1,

Equation

[0032] (2) When there is a place where the consecutive numerical values are arranged in the order of positive, negative, negative, positive (where one of the negative values may be 0), the corrected simple pure tone degree (PR’r) is obtained from the sound pressure level values with the test specimen in these four frequency bands based on the following formula 2,

Equation

[0033]

Number

[0034] The first simple evaluation method of the present invention will be described based on an example along the evaluation flow in the case of 1 / 3 octave band frequencies. The evaluation flow is shown in FIG. 4. Check the shape of the building on the production drawings, conduct an investigation to compare with the similarity to buildings measured in the past, measurement data in past buildings, and past materials, and evaluate the frequency of wind noise generated from the building itself. As a result of the evaluation, if there is no possibility of wind noise generation, the study is terminated. If there is a possibility of wind noise generation, a wind tunnel test is conducted.

[0035] Create a test body by modeling the building, and use a noise meter to measure the sound pressure level and noise level (L A) and measure them. Measuring without a test specimen is to remove the influence of noise derived from the blower of the wind tunnel device or the like. As the center frequency of the 1 / 3 octave band, for example, those defined in JIS C1513-1:2020 can be used. The results are shown in Table 4 and Figure 5. [Table 4]

[0036] For the obtained results, obtain the sound pressure level difference between with test specimen and without test specimen. Figure 6 shows the calculation results of the sound pressure level difference. If the calculated level difference exceeds the specified value in even one band, it is considered that wind noise may be generated, and subsequent investigations are carried out. If it is below the specified value at all frequencies, it is determined that no wind noise is generated, and the investigation ends. When the specified value is 5 dB, in this example, 250 Hz to 2.5 kHz and 4 kHz exceed the specified value, and it is determined that wind noise may be generated.

[0037] When there is a band exceeding the specified value, for all the sound pressure level values with test specimen, the arithmetic mean value (L k ) of the center frequencies (f k-1 , f k+1 ) above and below the center frequency (f ave_k =(L k+1 +L k-1 ) / 2) is calculated, and the value obtained by subtracting the sound pressure level value (L ave_k ) at the center frequency (f k ) from the arithmetic mean value (L k ) (ΔL k =L ave_k -L k ) is obtained as an integer rounded to the first decimal place.

[0038] (1) When there is no place where consecutive values (ΔL k ) are arranged in the order of positive, negative, negative, positive (however, one of the negatives may be 0) In the example shown in Table 3, the consecutive values (ΔL kSince there is no place where [[ID=]]), are arranged in the order of positive, negative, negative, positive (where one of the negatives may be 0), using the measurement results of the sound pressure level values with the test specimen, for all 1 / 3 octave band center frequencies except for both ends, the simple pure tone level (PR’) is calculated by the above-mentioned formula 1. For example, in Table 4, when calculating the simple pure tone level at 400 Hz, it is calculated as follows from the sound pressure level values at 400 Hz, 315 Hz, and 500 Hz (85, 69, and 77 respectively).

[0039] Simple pure tone level (PR’: 400 Hz) = 85 - 10 log 10 [(10^6.9 + 10^7.7) × 0.5] = 85 - 10 log 10 (29031002.9) = 85 - 75 = 10 (dB)

[0040] For the measurement results shown in Table 4, the simple pure tone level (PR’) is shown in Table 5 and Figure 7.

Table 5

[0041] (Determination of evaluation rank) In this example, the maximum value of the simple pure tone level (PR’ M ) is 10 dB, and the total value with the noise level with the test specimen (PR’ M + L A ) is 91 dB (10 + 81). Applying this value to the wind noise index shown in Table 3, it is determined that it corresponds to Rank 5. And since it exceeds the specified rank 3, countermeasure plans such as design changes to suppress the generation of wind noise are considered and fed back. Repeating this, if it becomes rank 2 or lower which is below the specified value, the examination is terminated assuming there is no problem with wind noise.

[0042] (2) Successive numerical values (ΔL k ) where they are arranged in the order of positive, negative, negative, positive (where one of the negatives may be 0). Hereinafter, the noise shown in Table 1 will be used as an example for explanation. Table 6 shows the arithmetic mean values of the sound pressure levels adjacent to each other vertically for the sound pressure levels excluding both ends shown in Table 1. Table 7 shows the numerical values (ΔL k ) obtained by subtracting the analysis values (Table 1) from the arithmetic mean values (Table 6). Also, Fig. 8 shows the result of overlapping Table 1 and Table 6 of the wind noise having a peak at 447 Hz.

[0043]

Table 6

Table 7

[0044] As shown in Table 7 and Fig. 8, for the wind noise having a peak at 447 Hz, there is a location where the numerical values (ΔL k ) obtained by subtracting the measured values from the arithmetic mean values are arranged in the order of positive, negative, negative, positive over the range of 315 Hz to 630 Hz. The location where this relationship exists is "a state where one peak is divided into the sound pressure level values L L and L U (where the frequency L < the frequency U)", and the composite value is obtained from the above formula 3, and L MS ' is taken as the center frequency for calculating the corrected simple phon.

[0045] In this case, from the values of 78 dB at 400 Hz (frequency L) and 500 Hz (frequency U) in Table 1,

Number

[0046] Next, the calculation of the corrected simple phon (PR’r) is performed. In the case of Table 1, L L-1 ’ is at a level of 64 dB at 315 Hz, and L U+1 ’ is at a level of 62 dB at 630 Hz. Substitute this into the above formula (2) and obtain as follows.

Equation

[0047] The corrected simple phonetic degree (PR’r) became 18 dB. The simple phonetic degree (PR’) obtained from the values in Table 1 without correction is 3.2 dB. (Determination of evaluation rank) According to the obtained simple phonetic degree and the corrected simple phonetic degree, the maximum value of the simple phonetic degree or the corrected simple phonetic degree (PR’ M ) is 18 dB, and the sum of this maximum value and the noise level with the test object (PR’ M + L A ) is 96 dB (18 + 78). When applied to the wind noise index shown in Table 3, it is evaluated as rank 4, and it is possible to avoid a determination on the dangerous side.

[0048] ·The second simple evaluation method The second simple evaluation method of the present invention is to define a wind noise index with the simple phonetic degree and the sum of the simple phonetic degree and the noise level as reference values, measure the sound pressure level value and the noise level (L A ) for each 1 / n octave band frequency (n is a natural number from 1 to 24) regarding the wind hitting sound for an actual building, for all the sound pressure level values, calculate the arithmetic mean value (L k ) of the center frequencies (f k-1 , f k+1 ) above and below the center frequency (f ave_k = (L k+1 + L k-1 ) / 2), subtract the sound pressure level value (L ave_k ) at the center frequency (f k ) from the arithmetic mean value (L k ) to obtain a numerical value (ΔL k = L ave_k-L k Find it as an integer obtained by rounding () to the first decimal place,

[0049] (1) When there is no place where consecutive numerical values (ΔL k ) are arranged in the order of positive, negative, negative, positive (where one of the negatives may be 0), calculate the simple pure tone degree (PR’) of the wind noise from the sound pressure level value based on the following formula 1,

Number

[0050] (2) When there is a place where consecutive numerical values (ΔL k ) are arranged in the order of positive, negative, negative, positive (where one of the negatives may be 0), obtain the corrected simple pure tone degree (PR’r) from the sound pressure level values with the test object in these four frequency bands based on the following formula 2,

Number

[0051]

Number

[0052] The evaluation flow in the case of the 1 / 3 octave band frequency in the second simple evaluation method of the present invention is shown in FIG. 9. The second simple evaluation method is the same as the first simple evaluation method, except that it does not conduct a wind tunnel test and evaluates only based on the sound pressure level values for each frequency of the 1 / 3 octave band of the wind hitting sound measured in an actual building and the noise level. By the second simple evaluation method, it is possible to evaluate existing buildings and the like for which a wind tunnel test cannot be performed.

Example

[0053] “Experiment 1” Regarding the method for calculating the corrected simple pure tone of the present invention, for 27 sets of data with the same peak prominence in cases (A) where the peak is near the center frequency of the 1 / 3 octave band and (B) where the peak is near the middle of two center frequencies and is divided into two bands by 1 / 3 octave band analysis, the correction effect was confirmed. The results are shown in Fig. 10. In Fig. 10, the horizontal axis represents the value of the simple pure tone level (PR’) calculated from the wind noise whose peak is near the center frequency of the 1 / 3 octave band (A), and the vertical axis represents the value of the simple pure tone level (PR’: white legend) or the corrected simple pure tone level (PR’r: light ink legend) calculated from the wind noise whose peak is near the middle of the two center frequencies (B).

[0054] In the case without correction, even if the simple pure tone level (PR’) of (A) increased, the simple pure tone level (PR’) of (B) did not increase, and the coefficient of determination regarding the correspondence of the entire 27 sets was as low as 0.21. On the other hand, when correction was performed, as the simple pure tone level (PR’) of (A) increased, the corrected simple pure tone level (PR’r) of (B) also increased, and the coefficient of determination regarding the correspondence of the entire 27 sets became very high at 0.90. Also, since there was no bias due to the frequency band, it was confirmed that this correction method is effective for a wide range of frequencies.

[0055] "Experiment 2" Using the horizontal continuous window unit curtain wall as the test specimen, wind noise was measured in a wind tunnel test chamber at wind speeds in the range of 2.5 to 25 m / s at intervals of 2.5 m / s, at wind direction angles in the range of 0 to 95° at a predetermined angle (0° when the front of the window is perpendicular to the wind direction), and with the test specimen present / absent. As an example, the spectrum at a wind speed of 25 m / s and a wind direction angle of 70° is shown in Fig. 11. As shown in Fig. 11, under these conditions, wind noise with a mountain-shaped peak was generated by the test specimen in the range of 300 to 1000 Hz. Note that the mountain-shaped peak generated in the range of 100 to 180 Hz occurred regardless of the presence or absence of the test specimen and was generated by the wind tunnel device.

[0056] "Reference Example" Table 8 shows the results of the evaluation of the wind noise with the test specimen by a skilled person.

Table 8

[0057] "Example 1" Regarding the wind noise with a test specimen, the method of the present invention using 1 / 3 octave band frequencies, that is, the maximum value of the simple sound purity (PR’ M ), and the sum of the maximum value of the simple sound purity and the noise level (PR’ M +L A ), the level of the wind noise was evaluated. The results are shown in Table 9.

Table 9

[0058] As shown in Table 9, the evaluation result by the simple evaluation method of the present invention was able to evaluate the wind noise at a wind speed of 25 m / s and a wind direction angle of 70° where a mountain-shaped peak occurred as Rank 5. By the simple evaluation method of the present invention, an evaluation result very similar to the evaluation result of an expert as a reference example was obtained.

[0059] 「Comparative Example 1」 Instead of the simple sound purity (PR’), TA described in Non-Patent Document 1 was used, and the value of the pure tone audibility (TA) and the sum of the pure tone audibility and the noise level (TA+L A ) were used to evaluate the level of the wind noise in the same manner as in Example 1 except that the level of the wind noise was evaluated from.

Table 10

[0060] The results are shown in Table 11.

Table 11

[0061] As shown in Table 11, the evaluation result using the pure tone audibility (TA) was significantly different from the evaluation result of an expert as a reference example. This is because the pure tone audibility (TA) could not detect a mountain-shaped peak with low pure tone, as the wind noise at a wind speed of 25 m / s and a wind direction angle of 70° where a mountain-shaped peak occurred was evaluated as Rank 0. From these results, it was confirmed that the method of the present invention can rank close to the evaluation results by experts and can reproduce the evaluation by experts at a high level.

Claims

1. Define a wind noise index using the simple pure tone level and the sum of the simple pure tone level and the noise level as reference values, Measure the sound pressure level values and the noise level for each 1 / n octave band frequency (n is a natural number from 1 to 24) with and without a test body in a wind tunnel test chamber. When the difference (with - without) in the sound pressure level values with and without the test body exceeds the specified value, For all sound pressure level values with the test specimen, the center frequency (f ｋ ) upper and lower center frequencies (f ｋ－１ , f ｋ＋１ ) arithmetic mean value (L ａｖｅ＿ｋ = (L ｋ＋１ +L ｋ－１ ) / 2) is calculated, The arithmetic mean value (L ａｖｅ＿ｋ ), the sound pressure level value (L ｋ ) at the center frequency (f ｋ ) is subtracted from the numerical value (ΔL ｋ = L ａｖｅ＿ｋ - L ｋ ), and the result is obtained as an integer rounded to the first decimal place. (1) When there is no place where the consecutive numerical values (ΔL ｋ ) are arranged in the order of positive, negative, negative, and positive (where one of the negatives may be 0), the simple pure tone level (PR') is obtained from the sound pressure level value with the test body based on the following formula 1. 【Number 1】 (However, L Ｍ ': Sound pressure level (dB) at the center frequency (fM) of the 1 / n octave band L Ｌ ': Sound pressure level (dB) at the 1 / n octave band center frequency (fL) of one of the low frequency bands of fM L Ｕ ’: Sound pressure level (dB) at the 1 / n octave band center frequency (fU) of one of the high-frequency bodies of fM (2) When there is a position where the consecutive numerical values (ΔL ｋ ) are arranged in the order of positive, negative, negative, positive (where one of the negatives may be 0), the corrected simple pure tone level (PR’r) is obtained from the sound pressure level values with the test specimen in these four frequency bands based on the following formula 2. 【Number 2】 (However, the center frequencies of the four frequency bands are in order f Ｌ－１ , f Ｌ , f Ｕ , f Ｕ＋１ and L ＭＳ ’: The composite value (dB) of the sound pressure levels at the 1 / n octave band center frequencies obtained by the following formula 3 is f Ｌ and f Ｕ ​ L Ｌ－１ ': Sound pressure level (dB) at the center frequency (f Ｌ－１ ) of the 1 / n octave band L Ｕ＋１ ': Sound pressure level (dB) at the center frequency (f Ｕ＋１ ) of the 1 / n octave band 【Mathematics 3】 L Ｌ : Sound pressure level (dB) at the center frequency (f Ｌ ) of the 1 / n octave band L Ｕ : Sound pressure level (dB) at the center frequency (f Ｕ ) of the 1 / n octave band The maximum value (PR') of the simple pure tone or the corrected simple pure tone Ｍ ), and the sum value (PR' Ｍ + L Ａ ) of the maximum value of the simple pure tone or the corrected simple pure tone and the noise level with the test object are ranked based on the wind noise index, and a simple evaluation method for building wind noise is characterized by this.

2. Define a wind noise index using the simple pure tone level and the sum of the simple pure tone level and the noise level as reference values, Measure the sound pressure level values and the noise level for each 1 / n octave band frequency (n is a natural number from 1 to 24) related to the wind hitting sound of an actual building, For all sound pressure level values, the arithmetic mean value (L ｋ =(L ｋ－１ +L ｋ＋１ )) of the center frequencies (f ａｖｅ＿ｋ =(L ｋ＋１ +L ｋ－１ )) above and below the center frequency (f ｋ )) is calculated, The arithmetic mean value (L ａｖｅ＿ｋ ), the sound pressure level value (L ｋ ) at the center frequency (f ｋ ) is subtracted from the arithmetic mean value (L ｋ ), and the resulting value (ΔL ａｖｅ＿ｋ = L ｋ - L ｋ ) is rounded to the nearest integer at the first decimal place, (1) When there is no place where the consecutive numerical values (ΔL ｋ ) are arranged in the order of positive, negative, negative, and positive (where one of the negatives may be 0), the simple pure tone level (PR') of the wind noise is calculated from the sound pressure level value based on the following formula 1, 【Number 4】 (However, L Ｍ ': Sound pressure level (dB) at the center frequency (fM) of the 1 / n octave band L Ｌ ': Sound pressure level (dB) at the 1 / n octave band center frequency (fL) of one of the low frequency bands of fM L Ｕ ': Sound pressure level (dB) at the 1 / n octave band center frequency (fU) of one of the high-frequency bodies of fM (2) When there is a location where the consecutive numerical values (ΔL ｋ ) are arranged in the order of positive, negative, negative, positive (where one of the negatives may be 0), the corrected simple pure tone level (PR’r) is obtained from the sound pressure level values with the test object in these four frequency bands based on the following formula 2. 【Number 5】 (However, the center frequencies of the four frequency bands are in order f Ｌ－１ , f Ｌ , f Ｕ , f Ｕ＋１ and L ＭＳ ': the composite value (dB) of the sound pressure levels at the 1 / n octave band center frequencies obtained by the following formula 3 is f Ｌ and f Ｕ ​ L Ｌ－１ ’: Sound pressure level (dB) at the center frequency (f Ｌ－１ ) of the 1 / n octave band L Ｕ＋１ ': Sound pressure level (dB) at the center frequency (f Ｕ＋１ ) of the 1 / n octave band 【Number 6】 L Ｌ : Sound pressure level (dB) at the center frequency (f Ｌ ) of the 1 / n octave band L Ｕ : Sound pressure level (dB) at the center frequency (f Ｕ ) of the 1 / n octave band The maximum value (PR') of the simple pure tone level or the corrected simple pure tone level Ｍ ), and the sum of the maximum value of the simple pure tone level or the corrected simple pure tone level and the noise level (PR' Ｍ + L Ａ ), are ranked based on the wind noise index, and a simple evaluation method for building wind noise is characterized by this.

3. The method for simple evaluation of building wind noise according to claim 1 or 2, characterized in that n is any one of 3, 6, 9, 12.

4. A building design method, characterized by feeding back the evaluation obtained by the method for simple evaluation of building wind noise according to claim 1 or 2, repeating the design of the building, and designing a building with reduced wind noise.

Images