JPH05333874A - Duct noise eliminating device - Google Patents

Abstract

PURPOSE:To obtain a device which is not affected by a sound outside a duct as an active sound eliminating device which detects a sound elimination state by a sound elimination state detector provided nearby a duct exit and provides the best sound elimination state at all times. CONSTITUTION:In the duct noise eliminating device equipped with a 1st sound detecting means 3 which detects time sound of a noise source 2, an acoustic source 4 for sound elimination, a sound elimination signal generating means 5 which generates a signal outputted from the acoustic source 4 for sound elimination to cancel a noise from the detection signal of the 1st sound detecting means 3 according to a specific generation pattern, a 2nd sound detecting means 6 which detects the sound elimination state, and a sound elimination pattern varying means 7 which varies the sound elimination signal generation pattern so that the sound detected by the 2nd sound detecting means 6 is minimized, the 2nd sound detecting means 6 is constituted having directivity so that the sensitivity to the output is low.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an active duct noise muffling device for reducing the emission of noise from a duct which is an air path to the outside of a device serving as a noise source, and particularly to a sound for detecting a muffling effect. The present invention relates to a duct noise muffling device that is provided with a detector near the outlet of a duct so as to always obtain the maximum muffling effect, and can sufficiently muffle the noise from the duct even if there is another sound outside. ..

[0002]

2. Description of the Related Art A large number of devices using air ducts such as ventilation of air conditioners into the air through ducts and ventilation of ducts for cooling the inside of various devices are used. FIG. 10 is a basic example of such a duct. In FIG. 10, a fan 102, which is a ventilation source, is attached to the back of the duct 101. The fan 102 serves as a noise source that generates noise as it operates, and the sound is transmitted to the outside through the duct 101 and becomes noise. If the device body generates noise, the sound is also transmitted to the outside through this duct.

Noise in offices, halls, etc. is a major problem, and various measures have been taken. However, noise transmitted through a duct cannot be blocked because the duct is an air path, and countermeasures are not easy. .. A widely used method for reducing duct noise is sound absorption processing by attaching a sound absorbing material inside the duct or sound absorption processing by a muffler. However, only relatively high frequency sounds can be silenced by these sound absorption processing, and in order to mitigate low frequency sounds, it is necessary to enlarge the sound absorbing material and muffler, and there is a problem that it is difficult to muffle low frequencies. .. Therefore, for low frequency sound, an active muffling device is used, which is provided with another sound source to muffle the sound by interference of the sound.

FIG. 11 is a diagram showing the basic construction of an active duct noise silencer. Reference numeral 111 is a duct, and a sound absorbing material is provided on the wall surface (not shown). As a result, almost all the high frequency sound propagating in the duct is absorbed, and only the relatively low frequency sound propagates to the outlet 118 of the duct 111, and it can be considered that the sound propagates in the duct 111 like a plane wave. 112 is a fan,
It becomes a noise source. Reference numeral 113 is a microphone that detects the sound from the noise source 112 and converts it into an electrical acoustic signal, and has a directivity that does not detect the sound from the exit 118. Reference numeral 114 is a muffling speaker. Reference numeral 115 denotes a muffling signal generating means, which uses a muffling speaker 114 based on a sound signal from the noise source 112 detected by the microphone 113.
A signal that cancels noise is generated at the exit 118 when output from the speaker 114 and output to the speaker 114. The muffling signal generation unit 115 has a transfer function when the sound from the noise source 112 propagates to the outlet 118 from the time when the sound is detected by the microphone 113, and a transfer function when the sound from the muffling speaker 114 propagates to the outlet 118. From the above, the filter has a generation pattern defined so that the two sounds become signals of opposite phases and the same intensity at the outlet 118.

In FIG. 11, a muffling speaker 114 is provided on the side surface of the duct 111, and the sound emitted from the muffling speaker 114 spreads as shown in the drawing, and becomes a plane wave in the duct direction at the exit 118, which is a noise source. Cancels with the sound from 112. Various proposals have been made for the position and the radiation direction of the muffling speaker. In the silencer shown in FIG. 11, the silence signal generation pattern is a fixed pattern that is predetermined based on the measurement result and the like. However, the propagation of sound in the duct and the sound quality of the noise source may change due to changes over time and changes in the environment. In such a case, it may happen that a fixed silencing signal generation pattern cannot provide a sufficient silencing effect.
Therefore, as shown in FIG. 12, in addition to the configuration of FIG. 11, a muffling effect confirmation microphone 126 is provided in the vicinity of the outlet 128 of the duct 121, and the muffling pattern varying unit 127 is operated according to the output thereof to confirm the microphone 126. The mute signal generation pattern is changed so that the output of is the lowest. As a result, a good silencing effect is always obtained.

The position of the muffling effect confirmation microphone 126 may be anywhere as long as the muffling effect can be confirmed.
For example, there is no noise between the position where the noise is canceled by the sound from the muffling speaker 124 and the exit, and the confirmation microphone 126 may be located anywhere within the range. However, for that reason, making the duct longer causes not only a problem that a large space is required, but also a problem that the ventilation efficiency is lowered due to an increase in resistance as an air path. Therefore duct 12
1 is as short as possible, and it is desirable that the confirmation microphone 126 is also provided near the outlet 128.

[0007]

Conventionally, it has been general to use an omnidirectional microphone having the same sensitivity in all directions as the muffling effect confirmation microphone 126. This is because not only the target noise and sound for muffling are limited to plane waves, but also sounds in different directions due to spherical waves and reflections are considered, and sound in all directions is not limited to a specific direction. This is because it was supposed that the sound could be muted when the two cancelled.

However, the microphone 126 for confirming the muffling effect
When an omnidirectional microphone is used as the above, if this microphone is installed near the outlet 128 for the above-mentioned reason, it is likely to be affected by the sound outside the duct 121.
Therefore, if the mute signal generation pattern is changed so that the output of this microphone becomes the lowest, the confirmation microphone 126 includes not only the sound in the duct but also the sound from the outside.
A sound that will be muted at the position will be output from the muting speaker 124. At this time, the sound from the noise source 122 emitted from the duct 121 and the sound from the muffling speaker 124 do not cancel each other, so that the outlet 1 of the duct 121 is
There is a problem that noise is emitted from 28.

In order to solve the above problems, the presence of an external sound is detected and the mute signal generation pattern is changed only when it is not present. However, a detector for detecting the presence of an external sound is separately provided. There is a problem that it needs to be provided. In addition, there is a problem that the silencing signal generation pattern cannot be modified indefinitely as long as the sound exists outside.

The present invention has been made in view of the above problems, and it is not necessary to provide a detector for detecting the presence of external sound, and the inside of the duct is always maintained even when external sound is present. The purpose of the present invention is to realize a duct noise muffling device that can appropriately muffle sound.

[0011]

FIG. 1 is a diagram showing the basic configuration of a duct noise silencer of the present invention. In the figure, 1
Is a duct that is an air path to the outside, and 2 is a duct 1
The duct noise muffling device of the present invention is a device for silencing noise emitted from the noise source 2 through the duct 1.

Reference numeral 3 is a first sound detecting means for detecting the sound of the noise source 2 and converting it into an electric signal. Reference numeral 4 denotes a sound deadening acoustic source that outputs a sound for deadening to the duct 1. Reference numeral 5 is a muffling signal generating means for generating a muffling signal from a sound signal of the noise source 2 detected by the first sound detecting means 3 in accordance with a predetermined generation pattern. When the muffling signal is output from the muffling sound source 4, the duct is provided. The sound from the noise source 2 propagating 1 is canceled. Reference numeral 6 denotes a second sound detecting means provided in the vicinity of the outlet 8 of the duct 1, and detects a sound in a state where the sound from the noise source 2 is muted by the sound from the sound deadening acoustic source 4. 7 is a silence signal generating means 5
Is a muffling pattern changing means for changing the generation pattern of the second sound detecting means 6 and minimizes the sound detected by the second sound detecting means 6. That is, it is changed so as to maximize the muffling effect. The duct noise muffling device comprises the above-mentioned first sound detecting means 3, a sound deadening acoustic source 4, a sound deadening signal generating means 5, a second sound detecting means 6,
The sound deadening pattern varying means 7 is provided so that the actual sound deadening effect is always optimized. However, in the present invention, the second sound detecting means 6 has directivity having low sensitivity to external sounds. Make something.

[0013]

When the sound deadening effect of the sound output from the duct 1 is detected by the second sound detecting means 6, it is necessary to detect the sound from the noise source 2 and the sound from the muffling speaker 4. Cannot come to the second sound detecting means 6 from the exit 8, and even if the second sound detecting means 6 has a directivity with low sensitivity in the direction of the exit 8, there is no problem. Absent. If the second sound detecting means 6 has low sensitivity in the direction of the outlet 8, the sound outside the duct 1 has a small effect on the second sound detecting means 6. Therefore, the sound detected by the second sound detecting means 6 is the duct 1
The noise output from the duct 1 can be minimized regardless of the presence of the sound outside the duct 1 by changing the sound to be detected to the minimum.

[0014]

EXAMPLE In the conventional active duct noise silencer shown in FIG. 12, the noise to be silenced is a sound whose frequency can be regarded as a plane wave whose wavelength is equal to or smaller than the diameter of the duct. If the noise propagating in the duct has a high frequency component, it is difficult to muffle the entire cross section of the duct, but as mentioned above, the high frequency sound can be sufficiently removed with a sound absorbing material, etc. I won't. However, even if there is high-frequency sound, it is possible to mute the sound from the muffling speaker at a specific point, and in such a case, the muffling confirmation microphone must not be affected by the sound outside the duct. However, the present invention can be applied.

However, in general, the active duct noise muffling device is targeted for low-frequency sound whose propagation in the duct can be regarded as a plane wave, and the embodiments described below also target such low-frequency sound. To do. In this case, as shown in FIG. 11, the muffling speaker and the duct outlet can be sufficiently separated from each other, and the sound from the muffling speaker can be muffled by forming a plane wave having an opposite phase near the outlet. If the diameter of the duct is sufficiently smaller than the wavelength of the generated noise, the sound can be silenced by the principle shown in FIG. In FIG. 2, 21 is a duct, 22 is a noise source, 24 is a muffling speaker, and 28 is an outlet of the duct 21. The lower diagram is an enlarged view of the part of the muffling speaker. Since the diameter of the duct 21 is sufficiently smaller than the wavelength of the propagating sound, the transmission time of the sound output from the muffling speaker 24 to the opposite surface can be ignored. Therefore, the sound can be silenced by outputting from the silencing speaker 24 a sound that causes the sparseness of the propagating noise, which is the opposite of the denseness.

At this time, in order to detect the muffling effect, the microphone 261 is connected to the muffling speaker 24 as shown in the lower part of the figure.
May be provided immediately below the sensor and the sensitivity thereof may be equal in the duct direction and the muffling speaker 24 direction. Microphone 2
If the output of 61 is zero, it means that the sound is completely muted. Of course, the sensitivity of the microphone 261 toward the outlet 28 is low. Since the sound is actually muted below the sound-deadening speaker 24, the same state can be detected even if the sound-deadening effect is detected at the position of the microphone 262 in the figure. In this case, the microphone has low sensitivity on the outlet 28 side and high unidirectionality in the opposite direction. However, it is not preferable to provide the microphone 262 in such a form because the duct must be lengthened accordingly, and therefore, in all the embodiments described below, the microphone 261 is provided directly below the muffling speaker 24.

The configuration of the first embodiment of the present invention is shown in FIG.
In FIG. 3, 31 is a duct, 32 is a noise source, and 33 is a first microphone that detects the sound of the noise source. The first microphone 33 has little sensitivity to the direction of the outlet 38. Reference numeral 34 is a muffling speaker, and 36 is a second microphone for detecting the muffling effect. Reference numeral 351 is an A / D converter that converts the output of the first microphone 33 into a digital signal. Reference numeral 352 is a digital signal processor (hereinafter referred to as DSP), which constitutes an adaptive filter. This will be described later. 353 converts the digital silencing signal generated by the adaptive filter into an analog signal and outputs it to the silencing speaker 34. 354 is a ROM, 355 is a RAM, and 3
Reference numeral 56 is a processor, which constitutes a computer. Reference numeral 371 is an A / D converter for converting the output of the second microphone 36 into a digital signal. Processor 3
56, RAM 355, ROM 354, DSP 352 and A / D converter 371 are connected to the bus, and the computer reads the output of the A / D converter 371, and the DSP 3
The adaptive filter formed by 52 is changed.

The operation of the duct noise silencer of the first embodiment shown in FIG. 3 will be briefly described. The sound from the noise source 32 is detected by the first microphone 33. The sound detected here is a part of the noise from the noise source 32, and the noise propagates as it is in the duct 31 toward the outlet 38. Therefore, the sound corresponding to the sound detected by the first microphone 33 reaches immediately below the muffling speaker 34 after a predetermined time. Therefore, the sounds that cancel each other based on the sound detected by the first microphone 33 at this time are muted. The noise can be silenced by outputting from the speaker 34. Therefore, the DSP 352 forms an adaptive filter for generating a muffling signal that produces such canceling sounds, the noise signal detected by the first microphone 33 is converted to a digital signal by the A / D converter 351, and then this adaptive filter is generated. Generates a silence signal at the D / A converter 35
After being converted into an analog signal in 3, the sound is output from the muffling speaker 34.

The second microphone 36 detects this mute state. If the output of the second microphone 36 is zero, it means that good muffling has been performed. Therefore, after the output of the second microphone 36 is converted into a digital signal by the A / D converter 371, it is read by the computer formed by the processor 356 or the like, and the coefficient of the adaptive filter formed by the DSP 352 is minimized so that this value is minimized. Change.

FIG. 4 shows an example of a muffling signal generating applied filter formed by the DSP 352, which is a well-known FIR filter (finish impulse response filter). 411 to 413 are delay devices, and there are n delay devices. 42
Reference numerals 1 to 424 denote multipliers, (n + 1) in number. 425
Is an adder. By inputting the signal Sin to this filter, the output signal Sout is obtained, which becomes the muffling signal. The characteristic of this filter is the coefficient h multiplied by each multiplier.
It can be changed by changing (0), h (1), h (2) ... h (n).

This change is performed by using the output of the second microphone 36 as an error signal and correcting each coefficient so that the root mean square of this error signal is minimized. This method is a well-known LMS (Least Mean Square Error) method, and since it is not directly related to the present invention, its detailed description will be omitted. As described above, the second microphone 36 does not have sensitivity in the direction of the outlet 38, and it is necessary that the second microphone 36 has the same sensitivity in the directions of the muffling speaker 34 and the noise source 32. One way to obtain a microphone with such characteristics is to select a suitable microphone from various commercially available directional microphones. As another method, as shown in FIG. 5, a shield plate 562 is provided on a part of the omnidirectional microphone 561.
Is provided to block sound from a specific direction. By directing the direction in which the shield plate 562 is provided to the outside, a directional microphone having low sensitivity in this direction and equal sensitivity in the direction of the noise reduction speaker and the noise source can be realized.

As another method for realizing a microphone having a predetermined directivity by using an omnidirectional microphone,
By synthesizing the outputs of a plurality of omnidirectional microphones in a predetermined positional relationship according to an appropriate synthesis pattern,
A method of realizing a desired directivity is known. Here, the microphone thus realized will be referred to as a synthetic microphone.

FIG. 6 is a diagram showing a synthetic microphone composed of two microphones. Reference numeral 661 is a synthetic first microphone, and 662 is a synthetic second microphone. 66
Reference numeral 3 is a first delay device, which delays the output of the synthetic first microphone 661 by a predetermined amount. Similarly, 664 is a second delay device that delays the output of the synthetic second microphone 662. Intensity converters 665 and 666 change the output intensities of the first delay device 663 and the second delay device 664, respectively. 667 is a subtractor. It is usually possible to dispense with one delay and the intensity converter.

A desired directivity can be realized by appropriately setting the distance d between the two microphones, the delay amount of the delay device and the intensity change amount of the intensity converter. For example, in FIG. 6, the two microphones have the same sensitivity and the second delay device 6
The delay amount of 64 is set to the time required for the sound wave to propagate through the interval d, the delay amount of the first delay device 663 is set to zero, and the conversion amount of the intensity converters 665 and 666 is set to 1, that is, no change. If so, there is no sensitivity to the right direction. In the upper part, the sensitivity is high for the sound of wavelength 2d, and the sensitivity is low for the sound of wavelength d. For the left side, high sensitivity is obtained when the wavelength is 4d. In any case, if the interval d and the target frequency are determined, various directivities can be realized by changing the delay amount and the intensity change amount.

The delay device, the intensity converter, and the subtractor can be realized by an analog circuit or a digital circuit. The second embodiment is an example in which a synthetic microphone that synthesizes the outputs of two microphones by digital processing is used as a muffling effect confirmation microphone. FIG. 7 is a diagram showing the configuration of the second embodiment. The difference from the first embodiment of FIG. 3 is that the second microphone 36 and the A / D converter 371 are changed to a synthetic microphone, and that the noise source 720 is controlled via the I / O port 357. Is.

The synthetic microphone is composed of two microphones 761 and 762, two A / D converters 771 and 772, and a DSP 773. In the DSP773, the delay device, the intensity converter and the subtractor shown in FIG. 6 are formed by a digital circuit, and the delay amount and the intensity change amount are processed by the processor 3.
It can be changed from 56. The synthetic microphone is set to have almost no sensitivity to the exit direction of the duct 710, but since it is installed near the exit, when the direction or frequency of the external sound changes, the synthetic microphone does not respond to the external sound. It can happen that they become sensitive. Therefore, the noise source 72 is transmitted through the I / O port 357 at an appropriate time.
The operation of 0 is stopped. At this time, only the external sound enters the synthesis microphone, and in that state, the coefficient of the digital circuit formed by the DSP 773 is previously stored in the ROM 35.
The pattern is changed according to a plurality of types of patterns stored in 4. Then, a pattern in which the output of the synthetic microphone is minimized is determined. The directivity of the synthetic microphone defined by the coefficient pattern of this digital circuit is the directivity with the lowest sensitivity to the actual external sound. After the setting of such a synthetic microphone is completed, the noise source 7
20 is operated, and the mute signal generation adaptive filter 752 is changed so that the output of the synthesis microphone becomes minimum.

The flowchart of FIG. 8 shows the processing by the computer of the setting operation for making the above-mentioned synthetic microphone the least sensitive to the actual external sound. This computer has a processor 356 and a RAM 355.
And a ROM 354. In step 801, the noise source 72 is output from the I / O port 357.
A signal for stopping the operation of 0 is output, and the silence speaker 740 is stopped in step 802. Mute speaker 740
Is stopped by turning off the power via the I / O port or fixing the input signal to zero. Step 80
In 3, the timekeeping process is performed to wait until the noise is sufficiently attenuated. In step 804, the type of coefficient pattern of the digital circuit formed by the DSP 773 stored in the ROM 354 is stored in the register i.

In step 805, it is determined whether the value of the register i is zero, and it is determined whether the test is completed for all the coefficient patterns. In step 806, the coefficient pattern corresponding to the order of the value of the register i is read from the ROM 354 and set in the DSP 773. Step 807
Then, the output of the synthetic microphone in the set state is read and the result is stored in the RAM 355. Step 8
At 08, the value of register i is decremented by 1 and the process returns to step 805.

After the outputs of all the coefficient patterns are stored as described above, the outputs of the synthetic microphones stored in step 809 are compared to determine the coefficient pattern having the smallest output. Is set to the DSP773, and the process ends. After that, the noise source 720 is operated again to return to the normal operation. In the second embodiment, the directivity to the outside is changed by changing the coefficient pattern of the synthetic microphone, but the sensitivity of the synthetic microphone to the noise reduction speaker 740 and the noise source 720 may be the same as shown in FIG. The plurality of coefficient patterns that are necessary and stored satisfy the above conditions even if the directivity to the outside changes.

However, in the case of a synthetic microphone using two omnidirectional microphones as shown in FIG. 6, the distance d between the microphones is fixed, and the condition that the noise source and the muffling speaker have the same sensitivity is satisfied. At times, if the directivity to the outside is changed, the sensitivity to the noise source and the muffling speaker changes, but the sensitivity changes. What is required for the synthetic microphone for confirming the muffling effect in the second embodiment is that the noise source 720 and the muffling speaker 940 have the same sensitivity and a high sensitivity as compared with the outside. That is sensitivity. That is, the ratio of the sensitivity to the noise source or the muffling speaker and the sensitivity to the outside is high. This is a general signal / noise ratio (S / N ratio)
Is equivalent to.

Therefore, in the third embodiment, a synthetic microphone capable of maximizing the S / N ratio is used. The configuration of the third embodiment is shown in FIG. The third embodiment is different from the second embodiment in that there are a large number of omnidirectional microphones 960 used in a synthetic microphone, and two suitable outputs can be selected by a switch 971 from them. The processing circuit of the microphone is an analog circuit. This analog processing circuit is composed of delay devices 972 and 973, such as BBD elements, whose delay amount is variable, variable resistors 974 and 975, and a differential amplifier 976 formed by an operational amplifier 976. The output is converted into a digital signal by the A / D converter 977, read by a computer including a processor 956, and the DSP 952 is read according to the value.
The adaptive filter coefficient of is changed. The switch 971, the delay devices 972 and 973, and the variable resistors 974 and 975 are controlled via the I / O port 957.

The positional relationship of the omnidirectional microphones used in the synthetic microphone is important, and by selecting a microphone to be combined from a plurality of microphones as shown in FIG. 9, a combination of microphones having different positional relationships can be selected. Instead of providing a plurality of microphones, a moving device that moves two omnidirectional microphones may be provided and moved to a desired position.

In the third embodiment, actually the most S / N
The operation of setting a synthetic microphone with a high ratio will be described. Various arrangements and combinations of microphones are conceivable, but here, as shown in FIG. 9, omnidirectional microphones are arranged in the direction of the duct 910, and the microphone groups at both ends and the next inner group are arranged. Two microphones are combined so that the distance between the microphones can be changed.

First, the noise source 920 and the muffling speaker 940.
And switch 971 selects the microphones at both ends. And the delay devices 972 and 9 are connected via the I / O port.
A predetermined pattern stored in the ROM 954 is set in the variable resistor 73 and the variable resistors 974 and 975, and the output thereof is stored.
There may be a plurality of patterns set at this time, and the respective outputs are stored. Do this for all microphone combinations. The value stored here corresponds to noise.

Next, the noise source 920 is operated with the muffling speaker 940 stopped, and the same operation as described above is performed for all microphone combinations. Since the value stored here corresponds to the sum of the signal and the noise, the ratio with the above noise is calculated for each combination, and the combination that maximizes the S / N ratio is determined. As described above, the process here is performed by performing steps 802 to 808 of the process flowchart of the second embodiment shown in FIG. 8 with the noise source stopped and in operation, and at step 809, the S / N ratio is set. This is a calculation, and the flowchart in the third embodiment is omitted.

As described above, in the present invention, a microphone having low sensitivity to the outside is used as the second microphone for confirming the muffling effect, but a system having a great flexibility can be realized by using the synthetic microphone.

[0037]

According to the present invention, the active duct noise muffling device for always optimally muting the duct noise by the sound sensor for confirming the muffling effect provided in the vicinity of the outlet according to the present invention, always reduces the duct noise regardless of the sound outside the duct. It will be able to mute properly.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic configuration of a duct noise silencer of the present invention.

FIG. 2 is a diagram showing a silencing principle in an embodiment of the present invention.

FIG. 3 is a diagram showing a configuration of a first embodiment.

FIG. 4 is an FIR example of an applied filter for silencing signal generation.
It is a figure which shows the circuit structure of a filter.

FIG. 5 is a diagram showing an example in which an omnidirectional microphone is provided with a shielding plate to give directivity.

FIG. 6 is a diagram showing a configuration of a synthetic microphone that synthesizes outputs of two microphones.

FIG. 7 is a diagram showing a configuration of a second embodiment.

FIG. 8 is a flowchart showing a setting processing operation of the synthesis microphone in the second embodiment.

FIG. 9 is a diagram showing a configuration of a third exemplary embodiment.

FIG. 10 is a diagram showing an example of a duct to which the present invention is applied.

FIG. 11 is a diagram showing a configuration example of an active duct noise silencer.

FIG. 12 is a diagram showing a configuration of a duct noise muffling device that changes a muffling signal while confirming a muffling effect and always maintains an optimal muffling state.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Duct 2 ... Noise source 3 ... 1st sound detection means 4 ... Silent sound source 5 ... Silent signal generation means 6 ... 2nd sound detection means 7 ... Silent pattern changing means 8 ... Exit

 ─────────────────────────────────────────────────── ─── Continuation of front page (72) Inventor Toshiro Oga 1015 Kamiodanaka, Nakahara-ku, Kawasaki-shi, Kanagawa Fujitsu Limited

Claims (5)

[Claims] 1. A duct noise muffling device for silencing noise from a noise source (2) emitted through a duct (1) which is an air path to the outside, which detects the sound of the noise source (2). One sound detecting means (3), and a sound deadening acoustic source (4) for outputting sound into the duct (1).
And a muffling signal for canceling the sound from the noise source (2) propagating in the duct (1) when output from the muffling sound source (4) of the first sound detecting means (3). A mute signal generating means (5) for generating a detected sound signal according to a predetermined generation pattern, and a second sound detecting means (6) provided near the outlet (8) of the duct (1) for detecting a mute state. ) And a muffling pattern varying means (7) for changing the generation pattern of the muffling signal generating means (5) so as to minimize the sound detected by the second sound detecting means (6). In a duct noise muffling apparatus for always optimizing a muffler effect, the second sound detecting means (6) has a directivity having a low sensitivity to a sound from the outside. apparatus. 2. The duct noise silencer according to claim 1, wherein the second sound detecting means (6) is a directional microphone. 3. The second sound detecting means (6) is a synthetic microphone including a plurality of omnidirectional microphones and a synthesizing means for processing outputs of the plurality of omnidirectional microphones according to a predetermined synthesis pattern. The duct noise silencer according to claim 1, wherein the duct noise silencer is provided. 4. The duct noise muffling apparatus according to claim 3, further comprising a synthetic pattern changing unit that changes the synthetic pattern of the synthetic microphone, wherein directivity of the synthetic microphone is variable. 5. The second sound detecting means (6) comprises a plurality of omnidirectional microphones, and a selecting means for selecting two outputs from the outputs of the plurality of omnidirectional microphones. The duct noise according to claim 1, further comprising a synthesizing unit that processes the two outputs in accordance with a synthesizing pattern corresponding to one output, and the synthesizing microphone has different directivities according to the two outputs to be selected. Silencer.