JPH0426898A - Active sound eliminating device - Google Patents

Abstract

PURPOSE:To eliminate a sound in a frequency range exceeding a 1st cross-mode frequency and to expand the frequency range of sound elimination by excluding a specific frequency range which includes a cross-mode frequency from an object area of sound elimination control intentionally. CONSTITUTION:This device is provided with an FIR filter which has a filter coefficient for generating an inverted sound signal whose phase is inverted as to the sound wave in a duct 1 according to the transfer function showing the propagation characteristic of the sound wave in the duct 1 and an additional sound source SS which receives the inverted sound signal of the FIR filter and radiates the inverted sound into the duct 1. The filter coefficient of the FIR filter is set to a filter coefficient regarding frequency bands above and below a specific frequency band except the specific frequency band including the cross-mode frequency of the duct 1. Sound elimination control is not performed intentionally in the specific frequency band including the cross-mode frequency, so the frequency band exceeding a conventional upper-limit frequency, i.e. primary cross-mode frequency becomes an object range of sound elimination control. Consequently, the controllable frequency band can be expanded.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an improvement in an active noise damping device that muffles noise in a propagation path such as a duct by applying inverted sound having an opposite phase to the noise.

(Prior art) Conventionally, in response to noise in a duct, an inverted sound having a phase opposite to that of the noise is emitted from an additional sound source such as a speaker. There are known devices that can also mute broadband noise (see, for example, Japanese Patent Application Laid-Open No. 61-296392, Japanese Patent Application Laid-Open No. 62-206212, etc.).

(Problem to be solved by the invention) By the way, when active noise reduction is performed in the duct as described above, the conditions under which the noise reduction is possible are that the change in sound pressure is one-dimensional only in the axial direction of the duct; In other words, the sound pressure has the same value within the same longitudinal section of the duct, and only under this condition can the inversion sound be added to the noise inside the duct.

For this reason, in the conventional system described above, the upper limit frequency at which the noise can be silenced is set to the frequency at which the above-mentioned silencing conditions do not hold, that is, the cross mode frequency determined by the diameter of the duct. For this reason, in a frequency band exceeding this upper limit frequency, control for silencing is not performed because it is necessary to consider the three-dimensional mode form.

The present invention has been made in view of the above, and an object thereof is to perform silencing control even in a frequency band exceeding the upper limit frequency of silencing control, thereby widening the frequency band in which silencing can be performed.

(Means for Solving the Problems) In order to achieve the above object, the present invention focuses on the fact that there are a plurality of duct cross-mode frequencies, and that the above-mentioned silencing conditions are satisfied in the frequency bands between them. .

For example, as shown in Figure 4, the cross mode frequency of a square duct when the inner wall is completely reflective is 0.2XO,
In a square duct of 2m (cross-sectional area 0, 4rr?), the primary crossmoto frequency is approximately 860Hz at the point indicated by symbol X.
, and the secondary frequency is approximately 1200H at the point indicated by symbol y.
z, and in the frequency band between the two, the sound can be muted under one-dimensional conditions.

Therefore, in the present invention, by intentionally excluding a predetermined frequency band including the cross mode frequency from the target of silencing control, the frequency band exceeding the first cross mode frequency can be subjected to silencing control. The frequency band will be expanded.

In other words, the specific solution of the present invention is an FIR filter having a filter coefficient that generates an inverted sound signal whose phase is inverted with respect to the sound wave in the duct based on a transfer function indicating the propagation characteristics of the sound wave in the duct. and an additional sound source that receives the inverted sound signal of the FIR filter and radiates the inverted sound into the duct. Then, a predetermined frequency band including the cross mode frequency of the duct was excluded from the filter coefficients of the FIR filter. The filter coefficients are set to target frequency bands before and after the predetermined frequency band.

(Function) With the above configuration, in the present invention, silencing control is not intentionally performed in a predetermined frequency band including the cross mode frequency, so silencing control is performed in a frequency band exceeding the conventional upper limit frequency, that is, the first cross mode frequency. become the target area of
The frequency band that can be controlled will be expanded accordingly.

(Effects of the Invention) As explained above, according to the active silencing device of the present invention, by intentionally excluding a predetermined frequency band including the cross mode frequency from the target area of silencing control, the primary cross mode frequency Since it becomes possible to mute sound in a frequency band that exceeds , it is possible to expand the frequency range that can be muted.

(Example) Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a schematic configuration of an active noise reduction device of the present invention, in which (1) is a duct as a propagation path for sound waves;
The shape of the duct (1) is, for example, 0.2xQ, 2m (
It has a square shape with a cross-sectional area of 0.4r#), and its inner wall is completely reflective. Therefore, the cross mode frequency of this duct (1) is theoretically the first cross mode frequency of about 860 Hz as shown in Figure 4 (in reality, there is a deviation from the theoretical value, for example, about 950 Hz). ),
Moreover, the secondary frequency is theoretically about 1200 Hz.

In addition, in the same figure, (m, n) (m, n-0~2) is y
This is the number of sound pressure nodes in each direction of the axis and y-axis. Further, FIG. 5 shows the cross mode frequency when the cross-sectional shape is circular.

Further, in FIG. 1, (NS) is a primary sound source placed at the left end of the duct (1) in the figure, which replaces a normal noise source.

Also, (Ml) and (M2) are the above-next sound source (NS)
Two detection microphones detect sound waves emitted from the duct (SS), an additional sound source consisting of a speaker placed from the center of the duct (1) to the right in the figure, and (No) the duct (1)
This microphone (Mo) is placed at the right end of the figure in 1), and this microphone (Mo) generates a composite sound wave of the sound wave emitted from the -order sound source (NS) and the sound wave emitted from the additional sound source (SS). Detect pressure level.

In addition, (SG) is a signal generator that emits sound waves from the -order sound source (NS), and (5) is the two detection microphones (
Ml) and (M2) are controllers that receive the detected sound pressure level of the primary sound source (NS) via a digital equalizer (DEQ). The controller (5) converts the sound pressure signal from the digital equalizer (DEQ) into A
FIR filter (which is received via a D converter (AD)
FIR). This FIR filter (Finite
1npulse Re5ponse Filter)
(FIR) has an impulse response with a finite duration, and is obtained based on the control flow shown in Figure 2 with respect to the sound pressure signal of the primary sound source (NS) received from the AD converter (AD). The phase is 180° from the filter coefficients
This generates an inverted inverted sound signal.

The inverted sound signal generated by the FIR filter (FIR) is output to the additional sound source (SS) via the built-in D/A converter (DA).
The sound wave emitted from the sound source (NS) is configured to be emitted as an inverted sound with a phase difference of 180@ from the sound wave of the -order sound source (NS).

In the figure, (FFT) is an FFT analyzer that analyzes the sound pressure level detected by the microphone (No.) placed at the right end of the duct (1).

Next, calculation of filter coefficients in the FIR filter (FIR) shown in FIG. 2 will be explained.

In FIG. 2, in step S1, filter coefficients are calculated based on a transfer function indicating the propagation characteristics of sound waves within the duct (1). Since this filter coefficient includes a predetermined frequency band including the cross mode frequency, in step S2, the above filter coefficient is converted into the frequency domain by Fourier transform.

Then, in step S3, a predetermined frequency band (for example, 90
0~1000H2), and the second cross mode frequency (
a predetermined frequency band (e.g. 11
The first and second A filter coefficient is obtained in which a predetermined frequency band including each of the cross mode frequencies is intentionally excluded from silencing control. Then, using the filter coefficients obtained in this manner, an inverted sound signal having a phase different by 180° from the sound wave of the primary sound source (NS) inputted from the digital equalizer (DEQ) is generated.

Therefore, in the above embodiment, as shown in FIG.
The frequency band of 100 to 1200 Hz is intentionally excluded from silencing control, so the conventional upper limit frequency (
900Hz) and below, as well as bands other than the above frequency bands.

That is, the band of l00CI-1100Hz, and 1200
In a frequency band exceeding Hz, the sound pressure level is lower in the example of the present invention shown by the broken line in the figure than in the case of only the sound pressure from the primary sound source (NS) shown by the solid line in the figure. Therefore, in this region, the sound pressure level cannot be reduced in the conventional example shown in Fig. 6, whereas the present invention can reduce the sound pressure level by a large amount as shown in Fig. 3 (b). It can be seen that it can be effectively reduced in the frequency domain.

In the above embodiment, the filter coefficients of the FIR filter (p; h) are fixed to a constant value, or the filter coefficients are updated sequentially using an adaptive FIR filter. Of course, you can.

[Brief explanation of drawings]

Figures 1 to 5 show embodiments of the present invention, with Figure 1 being a general schematic diagram, Figure 2 being a control flowchart for calculating filter coefficients, and Figure 3 being a diagram showing the reduction effect of silencing. 4 and 5 are diagrams showing cross mode frequencies, respectively. FIG. 6 is a diagram showing the reduction effect of conventional silencing. (1)...Duct, (SS)...Additional sound source, (No.
)...Microphone, (5)...Controller, (F
IR)...FIR filter. (1)...Duct (SS)...Additional sound source (Mo)...Microphone (5)...Controller (FIR)...FIR filter] ml) Fig. 4 Circular event appearance (ml) Fig. 4

Claims (2)

[Claims] (1) FIR having a filter coefficient that generates an inverted sound signal whose phase is inverted with respect to the sound wave in the duct (1) based on a transfer function indicating the propagation characteristics of the sound wave in the duct (1).
The FIR filter (FIR) includes a filter (FIR) and an additional sound source (SS) that receives an inverted sound signal of the FIR filter (FIR) and radiates the inverted sound into the duct (1).
The filter coefficient of the FIR) is set to a filter coefficient that targets a frequency band before and after the predetermined frequency band, excluding a predetermined frequency band including the cross mode frequency of the duct (1). active silencer. (2) The active noise reduction device according to claim (1), wherein the filter coefficients of the FIR filter (FIR) are set by Fourier transform and inverse Fourier transform.