JPH05224680A - Noise eliminating device for cooling device - Google Patents

Abstract

(57) [Abstract] [Purpose] In active control for silencing noise from a cooling device, it is determined that an abnormality has occurred in adaptive control for appropriately changing the calculation coefficient. [Arrangement] The arithmetic unit 16 processes the acoustic signal Sm from the vibration sensor 12 provided in the compressor 8 based on the acoustic transfer function to output the control signal Pa to the speaker 13.
The noise from the compressor 12 is muted by the control sound from the speaker 13. Adaptive control means 1
7 corrects the acoustic transfer function of the calculator 16. The determination unit 22 outputs impulse noise from the noise generator 19 so that the speaker 13 and the microphone 14 are
The impulse response function between them is obtained, and the abnormality of the adaptive control means 17 is determined based on the impulse response function.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a muffler used for a cooling device such as a refrigerator, and more particularly to a muffler for a cooling device which actively cancels noise from a machine room containing a compressor.

[0002]

2. Description of the Related Art In a cooling device using a compressor, for example, a refrigerator, it is often installed in a living room of a general household, and moreover, it is operated continuously regardless of the season. Noise reduction is one issue. In this case, the most problematic noise source of the refrigerator is noise from the machine room in which the compressor and the piping system connected to the compressor are stored. That is, in the machine room, the compressor itself generates comparatively large noise (operation noise of the compressor motor, fluid noise due to compressed gas, mechanical noise in the movable mechanical elements of the compression mechanism portion, etc.) and is connected to the compressor. The generated piping system also generates noise due to its vibration, and such machine room noise accounts for the majority of refrigerator noise. Therefore, suppressing the noise from the machine room greatly contributes to the noise reduction of the entire refrigerator.

Therefore, in the past, as a measure for reducing noise from the machine room, in addition to the noise reduction of the compressor itself (for example, the use of a rotary type compressor), improvement of the vibration isolation support structure of the compressor and the piping system. By improving the shape of the machine, the vibration can be damped in the vibration propagation path, or the sound absorbing member and sound insulating member are placed around the compressor and piping system to increase the sound absorption volume and noise in the machine room. The transmission loss is being increased.

However, in general, the machine room of the refrigerator is
Since it is necessary to release the heat generated by the driving of the compressor to the outside, openings for heat dissipation are provided at a plurality of locations, and noise leaks to the outside through these openings. Therefore, the conventional noise reduction measures as described above are naturally limited, and the noise level reduction effect can be expected to be only about 2 dB (A).

On the other hand, in recent years, along with the development of electronics application technology, especially acoustic data processing circuits and acoustic control technology, active noise control is performed by utilizing noise interference. The application of technology is drawing attention. That is, this active control basically converts a sound from a noise source into an electric signal by a sound receiver (for example, a microphone) provided at a specific position, and converts the electric signal into a signal processed by a calculator. By operating the control sound generator (for example, a speaker) based on this, an artificial sound having the same wavelength and the same amplitude as that of the original sound (sound from the noise source) at the control target point is generated from the sound generator. , The original sound is to be attenuated by causing the artificial sound and the original sound to interfere with each other.

Further, in order to realize such active control, it is necessary to correct the characteristic variation due to the secular change of the components constituting the signal system for the noise reduction and the characteristic variation due to the ambient temperature. Therefore, in practical use, the calculation coefficient (acoustic transfer function) of the arithmetic unit is corrected by following the fluctuation of the silencing ability. Auxiliary sound receiver (for example, microphone) that monitors the muffling effect of the sound generator, and if the result of monitoring by this auxiliary sound receiver is outside the predetermined allowable range, the calculation coefficient of the arithmetic unit is changed by a predetermined amount and It is also considered that a so-called adaptive control is provided in which control means is provided for performing the changing operation until the monitoring result falls within the allowable range, and thus the silencing ability during active control is always kept optimum.

[0007]

By the way, when a component such as a speaker or a microphone provided for adaptive control as described above fails, adaptive control cannot be reliably executed. In such a case, a diagnostic sound is generated from the speaker for a predetermined time, and when the output from the microphone is above a certain level, it can be determined that the speaker and the microphone are normal,
When the ambient noise is large, the output level from the microphone is high even if the speaker is out of order, and there is a risk of making an erroneous decision. Further, in order to avoid the ambient noise, the influence of the ambient noise can be avoided by setting a large diagnostic sound from the speaker, but this causes a drawback that the diagnostic sound itself becomes loud.

The present invention has been made in view of the above circumstances, and an object thereof is to actively cancel noise caused by driving a compressor by the operation of a control sounding device, and also based on a sound receiving result by an auxiliary sound receiving device. Cooling that can reliably detect abnormalities in the above-mentioned control sounding device or auxiliary sound receiving device without generating a large amount of noise in the surroundings, with a muffling function that appropriately changes the active control condition associated with driving the compressor. It is to provide a silencer for the device.

[0009]

SUMMARY OF THE INVENTION According to the present invention, a sound generated when a compressor housed in a machine room is driven is converted into an electric signal by a detection means, and the electric signal is processed by a calculator. The sound emitted to the outside from the machine room is actively canceled by operating the control sounder by using the control sounder, and an auxiliary device for monitoring the muffling effect of the control sounder at every predetermined time. When the result of monitoring by the sound receiver and the auxiliary sound receiver is out of the predetermined allowable range, the operation coefficient of the operation unit is changed by a predetermined amount and the changing operation is performed so that the monitor result is within the allowable range. In a silencer of a cooling device having an adaptive control means for performing until it is settled, a noise generator that outputs impulse-shaped white noise from the control sounder,
The impulse response characteristic between the control sounder and the auxiliary sound receiver is obtained based on the electrical signal corresponding to the white noise output from the auxiliary sound receiver, and the control is performed based on the impulse response characteristic. It is provided with a judging means for judging an abnormality of the sound generator or the auxiliary sound receiver.

[0010]

When the refrigeration system is operated by driving the compressor, noise is generated by driving the compressor. At this time, the sound from the compressor is converted into an electric signal by the detecting means, and the arithmetic unit operates the control sounder based on the signal obtained by processing the electric signal. As a result, the sound from the compressor is canceled by the interference between this and the artificial sound output from the control sound generator. Further, the sound deadening effect by such active control is monitored by the auxiliary sound receiver at every predetermined time, and when the monitoring result is out of the predetermined allowable range, the control means performs the calculation. The calculation coefficient (acoustic transfer function) of the device is changed by a predetermined amount, and accordingly, the silencing effect by the active control is changed to fall within the allowable range. Such a change operation of the calculation coefficient is performed until the result of monitoring by the auxiliary sound receiver falls within the allowable range, and thereby so-called adaptive control is performed in which the silencing ability during active control is always kept optimal.

When an abnormality occurs in the control sounding device or the auxiliary sound receiving device, the above-mentioned adaptive control cannot be reliably executed. Therefore, the abnormality is detected as follows. That is, the noise generator outputs impulse-shaped white noise from the auxiliary sound receiver at an appropriate timing. Then, the auxiliary sound receiver receives the white noise and outputs an electric signal corresponding to the white noise. Therefore, the judging means determines between the control sounding device and the auxiliary sound receiver based on the electric signal from the auxiliary sound receiver. Find the impulse response characteristics between them. At this time, when the control sounding device and the auxiliary sound receiving device are normal, the impulse response level has a characteristic of being attenuated over time. If so, it is determined that the control sound generator and the auxiliary sound receiver are normal.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is applied to a refrigerator will be described below with reference to the drawings. Figure 3 showing the overall configuration of the refrigerator
In FIG. 1, reference numeral 1 denotes a refrigerator main body which is a cooling device main body, and a freezer compartment 2, a refrigerating compartment 3 and a vegetable compartment 4 are provided in this order from the upper side. A refrigeration system is provided inside the refrigerator, and these components will be described. That is, 5 is a cooler arranged at the back of the freezer compartment, 6
Is a fan that directly supplies the cold air generated by the cooler 5 to the freezer compartment 2 and the refrigerator compartment 3. Reference numeral 7 denotes a machine room formed in the lower part of the rear side of the refrigerator main body 1, in which a rotary type compressor 8, a condenser pipe 9 and a defrosting water evaporator 10 using so-called ceramic fins are housed. ing. In the driving state of the compressor 8, the refrigerant from the compressor 8 is supplied to the cooler 5 through a refrigerant passage (not shown) to be cooled, and the fan 6 is driven to generate heat between the cooler 5 and the inside of the refrigerator. Exchange is to take place.

FIG. 4 (here, the condenser pipe 9
And the defrosting water evaporation device 10 is omitted), the machine room 7 has a rectangular opening only on its back surface.
It will be closed by. At this time, the peripheral edge of the machine room cover 11 is airtightly attached to the opening edge of the machine room 7, and the left edge in the figure is a vertically elongated elongated rectangular shape for heat dissipation. The opening 11a is formed. That is, when the machine room cover 11 is mounted, the machine room 7 is in a closed state except for the heat radiation opening 11a. The machine room cover 11 is formed of a material (for example, a metal such as iron) having excellent thermal conductivity and large sound transmission loss.

Further, in FIG. 4, reference numeral 12 denotes a vibration sensor which is a detecting means attached to the compressor 8 and is provided so as to convert the vibration sound of the compressor 8 which is a noise source into an electric signal. Reference numeral 13 is a loudspeaker which is a control sounding device disposed in the machine room 7, and is located near the heat dissipation opening 11a in the back wall of the machine room 7 (corresponding to the bottom wall of the refrigerator body 1). It is attached and supported in a buried state in the part. Reference numeral 14 denotes a microphone, which is an auxiliary sound receiver provided in the heat dissipation opening 11a, which receives noise from the compressor 8 and sound from the speaker 13 and monitors the sound deadening effect of the speaker 13. It has become.

Then, as shown in FIG. 1, the speaker 1
3 is operated by the control signal Pa obtained by processing the electric signal Sm from the vibration sensor 12 by the calculator 16 in the anti-phase sound generation circuit 15, and the processing of the electric signal as described above is performed. The following is a principle of silencing by active control as described below.

That is, the principle of silencing by active control will be schematically described with reference to FIG. 5. The sound generated by the compressor 8 which is a noise source is S1, the sound generated by the speaker 13 is S2, and the vibration sensor. Vibration sound detected by 12 is R1
, R2 is the sound received by the microphone 14 provided in the heat dissipation opening 11a which is the control target point, and T1 is the acoustic transfer function between the output of the sound and the input point as described above.
Assuming 1, T21, T12, and T22, the following equation holds as a 2-input 2-output system.

[0017]

[Equation 1] Therefore, the sound S2 to be produced by the speaker 13 is S2 = (-T12.R1 + T11.R2) / (T11.T) from the above equation.
22-T12 · T21), but in this case R2 = 0 can be set because the acoustic level at the heat dissipation opening 11a is aimed at zero. As a result, S2 = R1.T12 / (T12.T21-T11.T22). As can be understood from this formula, the heat dissipation opening 11
In order to reduce the sound R2 at a to zero, the sound R1 received by the vibration sensor 12 is processed into a sound S2 which is processed by applying a filter of F = T12 / (T12 · T21−T11 · T22) (1). Is generated from the speaker 13, it is possible to theoretically reduce the sound level at the heat dissipation opening 11a to zero, and the calculator 16 performs such sound processing (calculation) at high speed. However, the control signal Pa is applied to the speaker 13.

Now, in the above equation (1), F = G, T
When replaced with 12 = Gso, T21 = Gam, T11 = Gsm, T22 = Gao, G = Gso / (Gso.Gam-Gsm.Gao) (2). Here, the meanings of Gso, Gam, Gsm, and Gao are such that the subscripts at the front stage correspond to the input side and the subscripts at the rear stage correspond to the output side (response side). For example, Gam is the input signal to the speaker 13 at the input side. And the acoustic transfer function when the output signal from the microphone 14 is measured on the output side. When the noise from the compressor 8 is detected by the vibration sensor 12, the vibration sensor 12 does not receive the sound from the speaker 13, so Gam can be regarded as zero. Therefore, the above equation (2) becomes G = -Gso / (Gsm * Gao) (3). Here, since Gso / Gsm = Gmo, the above equation (3) is G = -Gmo / Gao (4). That is, the electric signal from the vibration sensor 12 has a G value represented by the above equation (4).
The sound level at the heat dissipation opening 11a can theoretically be set to zero by causing the speaker 13 to generate a sound processed by applying a filter corresponding to.

On the other hand, in the case of the refrigerator constructed as described above, the noise level generated in the machine room 7 in response to the driving of the compressor 8 is, as shown in FIG. The state is such that it has the property of increasing in the 5 kHz band. Of the noises corresponding to each of these bands, the noises on the high frequency side can be attenuated by the transmission loss in the machine room cover 11 and the like.
Since the sound can be easily silenced by installing an appropriate sound absorbing member inside the vibration sensor 12,
Active control of noise by the speaker 13 and the calculator 16
The target frequency may be 700 Hz or less.

Further, in the case of performing the active control of noise as described above, it is theoretically possible to control the noise in the machine room 7 so as to be a one-dimensional plane traveling wave. Also in the above, it becomes important to perform it easily and accurately. Therefore, in the present embodiment, of the three dimensions D, W, and H in the depth, width, and height directions of the machine room 7 shown in FIG. Set larger than D and H (specifically, W = 60
By setting 0 mm and D = H = 200 mm), the standing wave of the sound in the machine room 7 is constituted only in the first mode. That is, for example, when the machine room 7 is assumed to be a rectangular cavity, the following equation holds.

[0021]

[Equation 2] Here, f is the resonance frequency (Hz), Nx, Ny, Nz are the modes in the X, Y, and Z directions, and Lx, Ly, and Lz are the dimensions in the machine room 7 in the X, Y, and Z directions ( That is, D, W,
H) and C are sound speeds. Therefore, from the above equation, X, Y, Z
The frequencies fx, fy, f of the first standing wave in each direction
We can find z.

That is, as described above, the depth dimension D = 2
When the width dimension W = 600 mm and the height dimension H = 200 mm are set to 00 mm, the frequency fx of the first standing wave in the X direction is Ny = Nz = 0, and the speed of sound C = 340 m.
/ Sec,

[Equation 3] Similarly, the frequencies fy and fz of the first standing wave in the Y and Z directions are

[Equation 4] Becomes As a result, the target frequency (= 700H
z) Below, the standing wave of the noise in the machine room 7 is established only in the mode in the Y direction (width direction), and the noise in the machine room 7 is regarded as a one-dimensional plane traveling wave. be able to. Therefore, at the time of noise reduction by active control of noise using the speaker 13 or the like, theoretical treatment of the wavefront becomes easy, and the noise reduction control can be performed easily and accurately.

Now, referring to FIG. 1, the electric signal So from the microphone 14 is input to the adaptive control means 17 in the anti-phase sound generation circuit 15 and used for adaptive control. The principle of control will be described with reference to FIGS. 8 and 9.

Here, as shown in FIG. 9, the compressor 8
Noise from the S, vibration sound from the vibration sensor 12 to M, sound from the speaker 13 to A, acoustic signal from the microphone 14 to O, acoustic transfer function from the compressor 8 to the vibration sensor 12 from Gsm, compressor 8 from Microphone 14
Is Gso, and the acoustic transfer function Gao from the speaker 13 to the microphone 14 is Gao.

On the other hand, in FIG. 8, the speaker 13 is connected to the noise generator 19 via the switch 18.
The noise generator 19 outputs a predetermined level of white noise having a predetermined frequency bandwidth for a short time. Also, switch 1
8 is set so as to turn on at a predetermined timing when the compressor 8 is not driven, and the noise generator 19 outputs an impulse-shaped white noise signal to the speaker 13 when the compressor 8 is turned on. At this time, adaptive
The acoustic transfer function Gao is obtained by the filter 20.

The vibration sound signal M from the vibration sensor 12 is supplied to the second adaptive filter 21 after being multiplied by the acoustic transfer function Gao obtained by the first adaptive filter 20. There is. The second adaptive filter 21 has a vibration sound signal M · Gao from the vibration sensor 12 and a sound signal O from the microphone 14 which are multiplied by Gao by the sound transfer function Gao.
The acoustic transfer function G for active control and the latest acoustic transfer function Gnew obtained by this adaptive control based on the difference between
The difference ΔG with the difference ΔG is calculated as described later. In this case, the acoustic transfer function G is an initial setting value or a value previously obtained by adaptive control. The acoustic transfer function Gao is the value obtained this time by the first adaptive filter 20.

Considering the driving state of the compressor 8, M = S.Gsm (5) O = S.Gso + A.Gao (6) A = M.G (7) You can In addition, the microphone 14 from the vibration sensor 12 through the second adaptive filter 21.
Since the path leading to can be expressed by M · Gao · ΔG = O (8), expanding equation (8) above gives ΔG = O / (M · Gao) = (S · Gso + A · Gao) / (M ・ Gao) = (S ・ Gso + M ・ G ・ Gao) / (M ・ Gao) = (S ・ Gso) / (M ・ Gao) + (M ・ G ・ Gao) /
(M ・ Gao) = (S ・ Gso) / (S ・ Gsm ・ Gao) + G = (Gso / Gsm) / Gao + G Here, Gso / Gsm = Gmo can be regarded as ΔG = Gmo / Gao + G. .. Now, when the optimum acoustic transfer function is regarded as Gnew, −Gmo / Gao = Gnew, so ΔG = −Gnew
+ G, which can be expressed by Gnew = G-ΔG. Therefore, by changing the acoustic transfer function G to the acoustic transfer function Gnew and then executing active control based on the acoustic transfer function Gnew, it is possible to change the coefficient in real time and maintain the optimum volume.

On the other hand, as shown in FIG. 1, the anti-phase sound generation circuit 15 is adapted to receive a drive command for the compressor 8 (hereinafter referred to as the comp-on signal Sa).
During the output of the comp-on signal Sa, the compressor 8
Also, the fan 6 is driven.

As shown in FIG. 2, the judging means 22 is configured to include the adaptive filter 20, and the transfer function G obtained by the adaptive filter 20.
The impulse response function is obtained based on ao, and the failure determination is performed based on the impulse response function. In this case, the judging means 22 is provided with a failure lamp 23 as shown in FIG.
When it is determined that the failure does not hinder the mute control, the failure lamp 23 is displayed, and when it is determined that the failure hinders the mute control, the failure lamp 23 is turned on and the abnormality signal Sk is output.
To the anti-phase sound generation circuit 15. At this time, the anti-phase sound generation circuit 15 stops the active control due to the input of the abnormal signal Sk.

Now, the function of the anti-phase sound generating circuit 15, that is, the function of the calculator 16 and the adaptive control means 17 will be described below.

That is, when the temperature of the freezer compartment 2 rises and reaches a predetermined temperature, the compon signal Sa is output, and the compressor 8 is driven accordingly, so that the vibration sensor 12 attached to the compressor 8 vibrates. To do. Then, the computing unit 16 samples the vibration signal Sm (acoustic signal) from the vibration sensor 12 during the input period of the compon signal Sa, and processes the sampled acoustic signal based on the acoustic transfer function described above. After that, the control signal Pa based on the processing is output. As a result, the control signal Pa is given from the arithmetic unit 16 to the speaker 13, and the control sound is emitted from the speaker 13 in response to this.
Active control is performed in which the noise caused by driving the compressor 8 and the control sound from the speaker 13 interfere with each other in the heat dissipation opening 11a, and the sound level thereof is lowered.

At this time, the adaptive control means 17 samples the electric signal So from the microphone 14 to monitor the silencing effect by the active control described above. Then, when the monitoring result is within the predetermined allowable range, the active control is repeatedly executed. As a result, while the comp-on signal Sa is being input, the control signal Pa corresponding to the noise accompanying the driving of the compressor 8 is output to the speaker 13 and active control is performed in real time, so that the acoustic component from the compressor 8 is generated. Even if the noise fluctuates, the noise can be attenuated by following the fluctuation.

However, the result of monitoring the muffling effect is
If it is out of the allowable range, the adaptive control circuit 17 changes the calculation coefficient (transfer function) in the calculator 16 by a predetermined amount in the direction in which the muffling effect is enhanced. As a result, the output from the speaker 13 is output. Is adjusted so that the muffling effect of the artificial sound from the speaker 13 falls within the allowable range.

When an abnormality occurs in the speaker 13 or the microphone 14, the above-mentioned adaptive control becomes uncertain. Therefore, in this embodiment, the abnormality of the speaker 13 or the microphone 14 is determined as follows. I am trying.

That is, in FIG. 10 showing the operation of the judging means 22, the judging means 22 turns on the switch 18 at a predetermined timing when the compressor 8 is not driven (step S1). Then, an impulse-shaped white noise signal is output from the noise generator 19 to the speaker 13, so that the impulse-shaped white noise is emitted from the speaker 13 and reaches the microphone 14. As a result, the judging means 22 causes the adaptive filter 20 to
The acoustic transfer function Gao between the speaker 13 and the microphone 14 is calculated in accordance with, and the impulse response function is obtained based on the acoustic transfer function Gao.

That is, the impulse response function h (t) in the time domain corresponds to the transfer function (frequency response function) H (f) in the frequency domain, and these can be obtained by Fourier transform.

That is, the relationship between the input and the output in the frequency domain is as follows: X (f) = H (f) F (f) X (f): output F (f): input H (f): frequency response function ( Transfer function).

On the other hand, the relationship between the input and the output in the time domain is as follows: x (T) = h (t) * f (t) x (t): output f (t): input h (t): impulse response function *: Can be expressed by convolution integral.

The relationship between the impulse response function h (t) and the frequency response function H (f) is

[Equation 5] Becomes Therefore, the impulse response function can be obtained based on the acoustic transfer function Gao corresponding to the frequency response function H (f) obtained by the adaptive filter 20.

Subsequently, the judging means 22 calculates the power of the impulse response function in the predetermined period by dividing it into the first half period (X1) and the second half period (X2) (step S3).
At this time, when the speaker 13 or the microphone 14 is normal, the level of the impulse response function becomes large in the first half and attenuates in the second half to become small as shown in FIG. Therefore, in such a case, the judgment means 22 returns to the normal routine from steps S4 and S5.

On the other hand, when the speaker 13 or the microphone 14 is abnormal, the power X1 in the first half period of the impulse response function becomes less than or equal to the set value as shown in FIG. 11, so that the process proceeds from step S4 to step S6. Then, the failure lamp 23 is turned on. If the power X2 in the latter half period is equal to or greater than the set value due to the abnormality of the speaker 13 or the microphone 14, the fault lamp 23 is turned on and the abnormality signal Sk is output. At this time, the abnormal signal Sk is output because the anti-phase sound generating circuit 15 causes noise because the power X2 in the latter half period is equal to or larger than the set value because the adaptive control may cause noise. This is to stop active control.

According to the above-described embodiment, the speaker 13 involved in the adaptive control outputs the impulse-shaped white noise to detect the abnormality in the adaptive control based on the impulse response function between the speaker 13 and the microphone 14. Therefore, it is only necessary to output the diagnostic sound from the speaker 13 for a short time, and it is possible to reliably detect the abnormality due to the adaptive control while reducing the noise due to the diagnostic sound.

[0043]

As is apparent from the above description, according to the muffler of the cooling device of the present invention, the noise generator for outputting the impulse-shaped white noise from the control sounding device is provided, and the noise generator responds to the white noise. The impulse response characteristic between the control sounding device and the auxiliary sound receiving device on the basis of the electric signal and the judgment for judging the abnormality of the control sounding device or the auxiliary sound receiving device based on the impulse response property. Since the means is provided, the noise caused by driving the compressor is actively canceled by the operation of the control sounding device, and the active control condition associated with the compressor driving is appropriately changed based on the sound reception result by the auxiliary sound receiver. It is said that it is possible to reliably detect an abnormality of the above-mentioned control sounding device or auxiliary sound receiving device without generating a large noise in the surroundings. It exhibits an excellent effect.

[Brief description of drawings]

FIG. 1 is a schematic electrical configuration diagram showing an embodiment of the present invention.

FIG. 2 is a block diagram schematically showing the principle of adaptive control.

[Fig. 3] Vertical sectional view of the refrigerator

FIG. 4 is an exploded perspective view showing a state in which a cover is removed from a machine room of a refrigerator.

FIG. 5 is a schematic configuration diagram showing a silencing principle by active control.

FIG. 6 is a schematic perspective view showing dimensional relationships of machine rooms.

[Fig. 7] Noise level characteristic diagram

FIG. 8 is a block diagram for obtaining an acoustic transfer function in adaptive control.

FIG. 9 is a block diagram showing the operation of active control.

FIG. 10 is a flowchart showing the operation contents of the judging means.

FIG. 11 is a diagram showing an impulse response function in a normal state.

FIG. 12 is a diagram showing an impulse response function at the time of abnormality.

FIG. 13 is a diagram showing an impulse response function at the time of abnormality.

[Explanation of symbols]

1 is a refrigerator main body (cooling device), 7 is a machine room, 8 is a compressor, 11a is a machine room cover, 11a is a heat dissipation opening, 12 is a vibration sensor (detection means), 13 is a speaker (control sounder), Reference numeral 15 is a reverse-phase sound generation circuit, 16 is a calculator, 17 is an adaptive control circuit (control means), 19 is a noise generator, and 22 is a judgment means.

Claims (1)

[Claims] 1. A sound generated when a compressor housed in a machine room is driven is converted into an electric signal by a detecting means, and a control sounder is operated based on the signal processed by the arithmetic unit. In this way, the sound radiated from the machine room to the outside is actively canceled, and an auxiliary sound receiver for monitoring the muffling effect of the control sounder at every predetermined time and this auxiliary sound receiver. And an adaptive control means for changing the calculation coefficient of the arithmetic unit by a predetermined amount and performing the changing operation until the monitor result falls within the allowable range when the monitor result by the sound device is outside the predetermined allowable range. In a silencer of a cooling device provided with the noise generator that outputs impulse-shaped white noise from the control sounder, and the white noise output from the auxiliary sound receiver. The impulse response characteristic between the control sounding device and the auxiliary sound receiving device is obtained based on the electric signal according to the noise and the abnormality of the control sounding device or the auxiliary sound receiving device is determined based on the impulse response characteristic. A silencer for a cooling device, which is provided with a judging means for judging.