JP2017075780A - High frequency sound measurement device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect the maximum sound pressure level of a high-frequency sound in the vehicle interior, etc., of an electric vehicle with high accuracy.SOLUTION: A sound pressure level at a discretionary point is estimated by modeling the sound field of a high-frequency sound into a plane wave along a plurality of axes centering around the origin. A microphone array used in this algorithm has a frame 1 supported by the sound field, a total of 8 support arms 2 each extending from the frame 1 so that tips gather in a sound reception unit 4, and microphones 3 each supported by tips of the support arms 2. One microphone 3A is arranged at a position that is the origin of the sound reception unit 4, and the remaining microphones 3B-3H are arranged in three-dimensional form at equidistant positions centering around the origin, constituting 7 axes centering around the origin.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a high-frequency sound measuring apparatus that measures a maximum sound pressure level in a sound field of a high-frequency sound emitted from an inverter, for example.

  For example, in an electric vehicle or a hybrid type vehicle, it is known that a strong electric system including an inverter for controlling electric power and a motor / generator emit an unpleasant high-frequency sound. Many are high-frequency sounds with a short wavelength of 5 kHz or more. When these high-frequency sounds are radiated into a space with many parallel reflecting surfaces typified by glass, such as in the passenger compartment, reflections from multiple paths A strong interference sound field is formed by waves.

  In order to take measures against such high-frequency sound, it is necessary to quantitatively evaluate the maximum sound pressure level of the target high-frequency sound in the vehicle interior, particularly in the sound field near the passenger.

  However, a technique for measuring the maximum sound pressure level of high-frequency sound in a sound field such as in a passenger compartment has not been sufficiently established.

  In Patent Document 1, a microphone array in which a large number of microphones are arranged in a triangular lattice shape is used, and a sound pressure level distribution is visualized by a beam forming method based on a sound pressure level for each frequency band of each microphone obtained by a multichannel FET. This technique is disclosed.

JP 2006-308409 A

  Since the high frequency sound causes reflection even in a narrow surface, the technique of Patent Document 1 disturbs the sound field by the structure of the microphone array, and cannot be quantitatively evaluated in the sound field originally intended to be measured.

  An object of the present invention is to provide a high-frequency sound measuring device using a microphone array having a novel structure capable of measuring the maximum sound pressure level of high-frequency sound in a three-dimensional space based on a novel algorithm. .

This invention uses a microphone array as a high-frequency sound measuring device that measures the maximum sound pressure level of high-frequency sound in a sound field,
The microphone array is composed of a frame supported by a sound field and at least four support arms each extending from the frame so that each tip end is gathered in the sound receiving unit,
At the tip of the support arm, at least four microphones in total are arranged, one at the position serving as the origin of the sound receiving unit and the rest three-dimensionally centered on the origin.

  According to the present invention, a plurality of microphones arranged three-dimensionally around the origin of one microphone is configured in a three-dimensional manner with a plurality of axes centered on the origin, and the high-frequency sound in the three-dimensional space is Maximum sound pressure level can be measured.

The perspective view which shows one Example of a microphone array. The side view of this microphone array. The top view of a microphone array similarly. The perspective view which shows the structure of the flame | frame before microphone mounting | wearing. The top view which expands and shows one of the support arms. Sectional drawing along the AA line of FIG. The longitudinal cross-sectional view of a support arm. Sectional drawing along the BB line of FIG. Explanatory drawing of the sound field model expressed with a plane wave. Explanatory drawing of the equipment of the accuracy test about the maximum sound pressure level. The graph which shows the result of the precision test about the maximum sound pressure level. The graph which shows the result of having tested the influence on the precision by the number of axes. The graph which shows the result of having tested the influence on the precision by the number of microphone arrays. The graph which shows the result of the experiment applied to the actual vehicle.

  First, the maximum sound pressure estimation algorithm on which the present invention is based will be described.

  For example, when it is desired to obtain the maximum sound pressure level near the driver's ear position in the passenger compartment of an electric vehicle or the like, assuming a plurality of sound receiving points, each sound receiving point is compared with the distance from the sound source to the sound receiving point. If the distance between the two is sufficiently small, the sound field near the sound receiving point can be approximated by superposition of plane waves.

FIG. 9 shows the intensity of the plane wave and the sound pressure at the sound receiving point in the sound field model approximated by a plurality of plane waves. In FIG. 9, assuming three axes (A _{1 to} A ₃ ) centering on the origin P ₀ , the plane wave intensity along each axis (traveling waves S _{1 to} S ₃ , backward waves S ₁ ′ to S 6 shows a model in which all six waves of ₃ ′ are measured with four microphones (sound receiving points: P _{0 to} P ₃ ), but the number of axes (that is, the number of modeled plane waves and the number of microphones) is shown. Is optional. When there is no reflected sound from a direction where there is no sound source, it is possible to model with only a traveling wave from the sound source side.

  If the intensity of the plane wave is S, the sound pressure at each sound receiving point is expressed by the following equation (1).

Here, k is the wave number, r is the distance from the origin P ₀ to each sound receiving point, and i is an imaginary unit.

The sound pressure of the sound receiving point _{P 0, P 1, P 2} , P 3 shown in FIG. 9, S ₁ ~S _3, the use of S ₁ '~S _3', axes A _1, A _2, Regarding the direction of A ₃ , it can be expressed as the following equation (2).

  From the equation (2), the relationship of the following equation (3) is obtained.

  Here, “P”, “H”, and “S” are as follows.

  Thereby, “S” of each axial component can be obtained by the following equation (4).

Here, “H ⁺ ” is a pseudo inverse matrix of “H”.

  By substituting equation (4) into equation (3), the sound pressure at an arbitrary point can be obtained.

That is, the sound pressure at an arbitrary point can be estimated as a combination of plane waves using the sound pressure actually detected at the sound receiving points P ₀ , P ₁ , P ₂ and P ₃ including the origin P ₀ . Then, the sound pressure estimation calculation is executed for a number of points in the target sound field space, and the maximum sound pressure level in the sound field can be obtained by obtaining the maximum value among them. .

  For example, for a limited volume space near the driver's ear position in the passenger compartment, the sound pressure at each lattice point when divided into a three-dimensional fine lattice is obtained, and the maximum sound pressure level is Can be sought. This is merely a calculation that does not require sound pressure measurement at each lattice point, and the maximum sound pressure level can be obtained very easily.

  As can be easily understood from FIG. 9, in order to spatially configure at least three axes centered on the origin, at least four sound receiving points, that is, microphones are required.

  The high-frequency sound measuring apparatus of the present invention obtains the maximum sound pressure level by using the algorithm as described above. However, since the target high-frequency sound (4 kHz or more) has a short wavelength, it is sound even in a narrow surface. Will be reflected and disturb the sound field you want to measure. Further, the distance to each sound receiving point (microphone) with the origin at the center is preferably ¼ or less of the wavelength in order to establish the above algorithm. For example, in the case of a high frequency sound of 10 kHz, the wavelength is about 34 mm, so it is necessary to arrange a plurality of microphones close to a distance of about 5 to 8 mm, for example.

  1 to 3 show an embodiment of a microphone array used in the high-frequency sound measuring apparatus of the present invention that relies on the above algorithm. The microphone array includes a metal frame 1 supported by a sound field to be measured, a total of eight support arms 2 each having a thin metal rod shape supported by the frame 1, and the support arms 2. The total number of microphones 3 is respectively supported at the tip of each. The eight microphones 3 are gathered in the sound receiving unit 4 in the center of the frame 1, and the one microphone 3 </ b> A located at the origin and the remaining seven microphones 3 </ b> B to 3 </ b> H located around the origin, Is included. In FIG. 1 to FIG. 3, for convenience of explanation, three directions of X, Y, and Z orthogonal to each other are shown together, and “up” and “down” are shown along the Z direction of the figure. Although the word is used, this does not mean that the posture when the microphone array is used is limited to the posture shown in FIG.

  As shown in FIG. 4, the frame 1 forms a perfect circle centered on the origin and is positioned on the XY plane, and is positioned on the XZ plane so as to be orthogonal to the circular frame 6. And a semicircular frame 7 having a semicircular shape corresponding to the upper half of a perfect circle centered on the origin. These are formed in a thin band shape so as not to disturb the sound field, and are integrally joined to each other at two intersections 180 ° apart. The semicircular frame 7 has a pair of leg portions 7a and 7a extending in parallel further downward from the intersection with the circular frame 6, and a connecting portion 7b in which the ends of the pair of leg portions 7a and 7a are connected to each other. And this connection part 7b is integrally attached to the attachment part 8 which consists of metal pipes. The attachment portion 8 is used for attachment to a sound field for the entire microphone array, for example, a seat back side position in the vehicle interior. In one embodiment for high frequency sound of about 8 to 10 kHz, the width of the belt-shaped circular frame 6 in the Z direction and the width of the semicircular frame 7 in the Y direction is about 2.5 mm which is sufficiently small with respect to the wavelength. And the radial thickness is even smaller.

  The circular frame 6 and the semicircular frame 7 are each provided with a projection 10 having a rectangular block shape for supporting the base end of the support arm 2 at a predetermined position. The metal supporting arm 2 having a thin rod shape is inserted into a through hole provided in the radial direction of the protruding portion 10, and then fastened with a fixing screw 11 from the side surface of the protruding portion 10 to the frames 6 and 7. It is fixed.

  The support arms 2 whose base ends are fixed to the protrusions 10 respectively extend inward along the radial lines of the circular frame 6 or the semicircular frame 7, and the distal ends of the eight support arms 2 are received at the center. It gathers in the sound part 4. FIG. 4 shows the configuration of the frame 1 and the support arm 2 when the microphone 3 is not attached.

  The eight support arms 2 and the eight microphones 3 have the same configuration, although the arrangement with respect to the origin is different. 5 and 6 show an enlarged view of the support arm 2 and one of the microphones 3 attached to the tip thereof. 7 and 8 show only the support arm 2 to which the microphone 3 is not attached. As shown in these drawings, each support arm 2 has a base end portion 2a in a hollow cylindrical shape and a tip end portion 2b formed in a semi-cylindrical shape. The groove 13 has a semicircular cross section along the axial direction (longitudinal direction). In one embodiment, the support arm 2 is configured by cutting the tip portion 2b of the metal tube having a hollow portion corresponding to the concave groove 13 into a semicircular cross section. The support arm 2 is desirably as small as possible so as not to disturb the sound field. In one embodiment, the support arm 2 is composed of a metal tube having a diameter of about 1 mm.

  In addition, you may comprise the part 2a of the base end side of the support arm 2 in the shape of a solid cylinder.

  The microphone 3 is a small omnidirectional microphone similar to a microphone used in, for example, a hearing aid, and is configured in a cylindrical shape as a whole. A total of three lead wires for power and signals from the center of the base end surface 3a. 15 (see FIG. 6) is pulled out. In one embodiment, the cylindrical axial dimension and diameter are both on the order of 2 mm. In the microphone 3, the base end surface 3 a is brought into contact with the distal end surface of the support arm 2, the lead wire 15 is disposed along the concave groove 13 of the support arm 2, and the support arm 2. The heat shrinkable tube 16 is placed on the distal end portion 2b and heated and shrunk to fix and support the tip of the support arm 2. That is, the microphone 3 is supported by the support arm 2 via the lead wire 15 drawn from the base end surface 3a. Since the used microphone 3 is extremely light, a practically sufficient support strength can be obtained by fixing the lead wire 15 with the heat shrinkable tube 16. In a state where the lead wire 15 is tightened in the radial direction by the heat shrinkable tube 16, at least a part of the three lead wires 15 is accommodated in the concave groove 13 as shown in FIG. It is located substantially at the center of the outer circle of the proximal portion 2a. As a result, the central axis of the cylindrical microphone 3 supported at the distal end of the support arm 2 and the central axis of the base portion of the support arm 2 supported by the frame 1, that is, the proximal end portion 2a, are substantially equal. I'm doing it.

  The tip surface 3b of each microphone 3 forms a circular surface orthogonal to the central axis of the support arm 2. In this embodiment, the center point of the tip surface 3b is regarded as the sound receiving point of each microphone 3. ing.

  Therefore, strictly speaking, the microphone 3A located at the origin is positioned at a position where the center of the tip surface 3b is the origin of the XYZ space, as shown in FIGS. As described above, the circular frame 6 and the semicircular frame 7 are formed in a circular or semicircular shape centered on the origin, and each support arm 2 is arranged along a radial line centered on the origin. .

  The other seven microphones 3B to 3H are three-dimensionally arranged at equidistant positions with the origin as the center so as to define the plane wave axis in the algorithm described above with respect to one microphone 3A serving as the origin. Yes. Regarding the microphones 3B to 3H, the center point of each tip surface 3b is regarded as a sound receiving point, and therefore the tip surfaces 3b of the microphones 3B to 3H are arranged so as to be equidistant from the origin. Yes. In one embodiment, the distance from the origin to the tip surface 3b of each of the microphones 3B to 3H is set to 5 mm, which is shorter than ¼ of the wavelength of the high frequency sound of 10 kHz.

  The specific arrangement of the microphones 3A to 3H and the respective support arms 2 will be described with reference to FIGS. 1 to 3. The support arm 2 supporting the microphone 3A for the origin is supported by the circular frame 6 and is XY. It extends in the Y direction on the plane. Accordingly, the tip surface 3b of the microphone 3A serving as the origin is along the XZ plane.

  The microphones 3B, 3C, 3D, and 3E are respectively supported by the circular frame 6 like the microphone 3A for the origin, and are arranged on the XY plane. When the XY plane is viewed as shown in FIG. 3, the microphones 3B, 3C, 3D, and 3E are arranged so as to have an angular difference of 45 ° with the origin as the center. In FIG. 3, the microphone 3 </ b> B is hidden by overlapping with the semicircular frame 7, and the support arm 2 is supported at the intersection of the circular frame 6 and the semicircular frame 7. Therefore, on the XY plane, with the origin at the center, one axis along the X direction is configured by the microphone 3B, and one axis along the Y direction is configured by the microphone 3D. The microphone 3C and the microphone 3E constitute two axes inclined at 45 ° with respect to the X and Y directions.

  The origin microphone 3A and the microphone 3B have an angle difference of 90 °, but this is a problem in the layout of the support arm 2 and is not an important angle in the above-described algorithm.

  The microphones 3F, 3G, 3H are respectively supported by the semicircular frame 7, and are arranged on the XZ plane together with the microphone 3B. When the XZ plane is viewed as shown in FIG. 2, the microphones 3B, 3F, 3G, and 3H are arranged so as to have an angular difference of 45 ° with the origin as the center. Note that the microphone 3B is hidden by overlapping with the circular frame 6 in FIG. Therefore, on the XZ plane, with the origin as the center, the microphone 3B constitutes one axis along the X direction described above, and the microphone 3G constitutes one axis along the Z direction. The microphones 3F and 3H constitute two axes inclined by 45 ° with respect to the X and Z directions.

  Since the tip surfaces 3b of the microphones 3B to 3H are all orthogonal to the central axis of the support arm 2, the tip surfaces 3b of the microphones 3B to 3H arranged as described above are all directed toward the origin. doing.

  Accordingly, when the arrangement of the microphones 3B to 3H is considered in three dimensions, the tip surfaces 3b of the seven microphones 3B to 3H are arranged along the surface of a sphere centering on the origin (in one embodiment, a sphere having a diameter of 10 mm). It becomes the shape that was made. The seven axes are three-dimensionally configured with the origin at the center. That is, seven axes including three axes in the X, Y, and Z directions and four axes having an angle of 45 ° with respect to two of these three axes are configured. As a result, it is possible to model the plane wave along the seven axes and accurately obtain the sound pressure level at an arbitrary point, and hence the maximum sound pressure level in the target sound field, according to the algorithm described above.

  In the microphone array configured as described above, the microphone 3A for the origin exists inside the sphere serving as the sound receiving unit 4, but the other microphones 3B to 3H are positioned on the outside of the sphere together with the respective support arms 2. To do. Therefore, there is little disturbance of the sound field due to reflection of high-frequency sound having a short wavelength. Moreover, since the tip surfaces 3b of the microphones 3B to 3H are along the surface of the sphere, the microphones 3B to 3H do not partially enter the inside of the sphere. In addition, the frame 1 and the support arm 2 are sufficiently thin compared to the wavelength, and are separated from the central sound receiving unit 4 to the outside, so that there is little disturbance in the sound field due to these.

  Although it is advantageous in terms of measurement accuracy and the like that the distance between the microphone 3A for origin and the other microphones 3B to 3H is the same for all axes as in the microphone array of the above embodiment, The present invention is not limited to the equidistant distance between the microphones on each axis.

  Furthermore, in the microphone array of the above embodiment, the inner microphone 3 and the support arm 2 are protected by the circular frame 6 and the semicircular frame 7 having relatively high strength and rigidity. For this reason, when actually handling the vehicle in the passenger compartment or the like, the position of the microphone 3 or the like will not be accidentally touched. As described above, the microphone 3 is supported by the support arm 2 with the lead wire 15 and the heat shrinkable tube 16, but is surrounded by the circular frame 6 and the semicircular frame 7 and is not subjected to external force from the outside. Sufficient durability is obtained.

  10 and 11 show a test for verifying the accuracy of the maximum sound pressure level obtained by the above algorithm using the above microphone array. FIG. 10 shows the equipment used for the test. In the anechoic chamber 21, a reference microphone 22 is arranged as shown in the figure, and the reference microphone 22 is placed on both sides at an equal distance (1 m). Speakers 23 and 23 are arranged to face each other. Then, the phase error and output error of the speakers 23 and 23 were corrected with values obtained by impulse response measurement, and the position of the reference microphone 22 was adjusted to the maximum sound pressure. On the other hand, the microphone array of the above-described embodiment is disposed in the vicinity of the reference microphone 22.

  In the test, a pure sound of 2000 to 10000 Hz (1/3 octave) is reproduced in the same phase from both speakers 23 and 23, and a measurement value by the reference microphone 22 is obtained as a reference maximum sound pressure level, and the microphone array of the embodiment is used. The maximum sound pressure level was estimated by the algorithm described above.

  FIG. 11 shows the error between the reference value by the reference microphone 22 and the estimated value by the microphone array of the embodiment and the algorithm described above for each frequency. As shown in FIG. 11, the error between the reference value and the estimated value by the reference microphone 22 is −0.6 to 1.7 dB. By using the microphone array of the embodiment, the maximum sound pressure level in the reflected sound field is obtained. It was proved that can be estimated accurately.

  In the microphone array of the above embodiment, seven axes centered on the origin are constituted by eight microphones 3. However, in the present invention, if three axes are constituted by at least four microphones, a three-dimensional space can be obtained. The maximum sound pressure level can be estimated.

  FIG. 12 shows the result of obtaining the relationship between the number of axes and the estimation accuracy of the maximum sound pressure level by simulation. This assumes five high-frequency sound sources of 8000 Hz, treats the arrangement, strength, and phase of the sound sources as parameters, and uses the calculated maximum sound pressure values for each condition as a reference, and the maximum sound pressure estimated by the algorithm described above. Is the evaluation function. A total of 9000 trials were calculated by randomly selecting the position of each sound source at a distance of 1 m from the center of the microphone array under three conditions for the combination of phases of the sound sources and four conditions for the combinations of strengths.

  As shown in FIG. 12, the maximum sound pressure can be estimated with sufficient accuracy if there are at least three axes, and very high accuracy can be obtained if there are seven axes as in the microphone array of the above embodiment. .

  The microphone array of the above embodiment can estimate the maximum sound pressure level by arranging one microphone array in a sound field such as in a vehicle interior, etc., but further increases the estimation accuracy by using a plurality of microphone arrays. Is also possible.

  FIG. 13 shows the result of obtaining the estimation accuracy of the maximum sound pressure level when the number of microphone arrays is increased by the same method as in FIG. As shown in the figure, the maximum sound pressure can be estimated with sufficient accuracy with one microphone array, and if four or more microphone arrays can be arranged, very high accuracy can be obtained. In addition, since the microphone array of the said Example can be comprised about about 10 cm in diameter, it is easy to arrange several in the vehicle interior of an electric vehicle etc., for example.

  FIG. 14 shows the result of verifying the accuracy of the present invention in an actual running experiment using an actual electric vehicle. In this test, three types of actual vehicles with differences in countermeasures against high-frequency high-frequency abnormal noise were prepared, and the maximum sound pressure level of each specification was estimated using the microphone array of the above example and the algorithm described above. On the other hand, for each specification, a sensory evaluation of the maximum sound pressure level by a skilled designated evaluator was performed, and as a conventional method, one microphone was placed near the driver's ear and the maximum sound pressure level was measured. FIG. 14 summarizes these three types for each specification.

  As shown in FIG. 14, the measurement of the sound pressure level using a single microphone, which is a conventional method, cannot correctly grasp the maximum sound pressure level. On the other hand, the estimated value of the maximum sound pressure level according to the present invention shows a tendency consistent with the sensory evaluation by a skilled designated evaluator.

  As described above, the present invention has been described mainly with respect to high-frequency sound in the passenger compartment by an inverter or the like in an electric vehicle or a hybrid vehicle. However, the present invention is not limited to such an application, and the high-frequency sound in various fields. It can be widely applied to sound measurement. Further, the target frequency is not limited to the above-described 8 to 10 kHz. For example, for a high frequency sound of 1 kHz or more, it is difficult to detect the maximum sound pressure level by a conventionally known method, and the present invention is useful for such a high frequency sound.

DESCRIPTION OF SYMBOLS 1 ... Frame 2 ... Support arm 3 ... Microphone 4 ... Sound receiving part 6 ... Circular frame 7 ... Semi-circular frame 8 ... Mounting part 13 ... Groove 15 ... Lead wire 16 ... Heat shrinkable tube

Claims (9)

A high-frequency sound measuring device that measures the maximum sound pressure level of high-frequency sound in a sound field using a microphone array,
The microphone array is
A frame supported by the sound field;
At least four support arms respectively supported by the frame and extending from the frame so that the tip ends of the frames are collected at the sound receiving unit;
At least four microphones, each supported at the tip of the support arm, one disposed at a position serving as the origin of the sound receiving unit, and the rest disposed three-dimensionally around the origin;
A high-frequency sound measuring device comprising: Modeled as a plane wave traveling along at least three axes defined by the microphone at the origin and other individual microphones;
The maximum sound pressure level in the sound field is obtained using an algorithm for calculating the intensity of the plane wave along each axis at an arbitrary point in the sound field based on the sound pressure measured by each microphone. Item 2. The high-frequency sound measuring device according to Item 1.   The high-frequency sound measuring apparatus according to claim 1 or 2, wherein the remaining microphones other than the microphone arranged at the origin are arranged at equal distances from the origin. The tip surface of one microphone is located at the origin, and
The high-frequency sound measuring device according to any one of claims 1 to 3, wherein the tip surfaces of the remaining microphones are arranged so as to be directed toward the origin. The tip of the support arm is formed in a semi-cylindrical shape with a concave groove along the longitudinal direction of the support arm,
The high-frequency sound measuring device according to claim 1, wherein a lead wire of a microphone disposed at a tip of the support arm is disposed along the concave groove. The base of the support arm has a cylindrical or columnar shape that is continuous with the semi-cylindrical surface of the tip,
6. The high-frequency sound measuring device according to claim 5, wherein the central axis of the support arm coincides with the central axis of the cylindrical microphone. The frame includes a circular frame centered on the sound receiving portion, and a semicircular frame along a plane orthogonal to the circular frame,
The high-frequency sound measuring device according to claim 1, wherein each support arm extends inward along a radial line from the circular frame and the semicircular frame.   The high-frequency sound measuring apparatus according to claim 7, wherein the frame includes a mounting portion that is continuous with the semicircular frame.   A total of eight axes are formed so as to constitute a total of seven axes including three axes orthogonal to each other centering on the origin and four axes having an angle of 45 ° with respect to two of the three axes. The high-frequency sound measuring apparatus according to claim 1, further comprising a microphone.

Images