JP2009237719A - Feedback type active muffler and vending machine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress the output of a high frequency control sound when an A/D converter for A/D-converting a signal output from a microphone is saturated. <P>SOLUTION: A feedback type active muffler having the microphone 24 for detecting a noise, a speaker 23 for outputting a control sound for canceling the noise, the A/D converter 32 for A/D-converting the output of the microphone 24 to generate a detection signal ej, a reference signal generation part 102 for removing the components of the control sound from the detection signal ej to compute a reference signal xj and a noise control filter 103 for generating a control sound signal uj for driving the speaker 23 from the reference signal xj includes a noise level determination part 106 for comparing the detection signal ej with a threshold a1 not higher than the saturation level of the A/D converter 32 to generate a determination signal, and an output control part 107 for suppressing the output of the control sound according to the determination signal. This suppresses the output of an unnecessary control sound when a loud noise is detected. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a feedback-type active silencer and a vending machine using the same. More specifically, the present invention relates to an improvement of an active silencer that generates a control sound for canceling noise by feedback control. In particular, the present invention relates to an improvement of a feedback type active silencer suitable for noise suppression of a compressor of a vending machine.

  2. Description of the Related Art An active silencer (ANC: Active Noise Controller) that detects noise with a microphone and cancels noise by outputting a control sound having the same amplitude and opposite phase from a speaker is conventionally known. Such active silencer control methods are classified into feed forward type and feedback type. However, the feedback type is difficult to adjust to achieve sufficient silence, and most of the conventional active silencer is feed forward type. The type is adopted.

  A feedforward type active silencer is a device in which a microphone (reference microphone) disposed upstream of a noise propagation path detects noise, and a speaker disposed downstream outputs a control sound to cancel the noise. That is, in the feedforward type, the noise itself detected by the microphone as a reference signal is canceled by the control sound from the speaker. For this reason, there is an advantage that a relatively good silencing effect can be obtained, but there is a disadvantage that noise can propagate only in one direction and can be applied only to a propagation path that can secure a sufficient distance between the microphone and the speaker. It was.

  On the other hand, the feedback type active silencer detects an environmental sound including noise and control sound as an error signal of a microphone (error microphone), and removes a control sound component included in the error signal by calculation to obtain a reference signal. The control sound is determined based on the reference signal. For this reason, there is an advantage that it can be applied regardless of the propagation path of noise, but there is a disadvantage that a silencing effect cannot be obtained unless it is periodic noise.

  Further, an active silencer using an adaptive filter as a filter that generates control sound from a reference signal has been proposed (for example, Patent Document 1). In such an adaptive active silencer, the transmission coefficient of the filter that converts the reference signal into the control sound signal is automatically updated, so that it can be learned to cope with changes in the noise source and propagation characteristics. In such an adaptive active silencer, the filter function is realized as digital processing using a DSP (Digital Signal Processor), and the output of the microphone is input to the DSP after A / D conversion.

  In general, vending machines that sell canned beverages have built-in temperature control compressors and cooling fans.When installed in a quiet environment, these built-in devices are used as noise sources. Can be. Therefore, it is conceivable to suppress the noise of the built-in equipment by installing an active silencer. The noise from these built-in devices varies depending on the driving state, but periodic noise is generated during a certain period of the driving state. Therefore, feedback-type active noise reduction can be applied to suppress such noise, and if an adaptive filter is used, it is possible to cope with changes in driving conditions and changes in the surrounding environment. It is thought that a silencing effect can be obtained.

  However, the microphone detects not only the noise from the built-in device but also the noise at the time of product purchase and product replenishment, and the noise from outside the vending machine such as human speech. For example, when a product is purchased, a large noise is generated when a canned beverage falls in the vending machine, and when the product is replenished, a large noise is generated by opening and closing the door or inserting the product. Since the intensity of such noise is much larger than the noise generated by a compressor, the A / D converter that performs A / D conversion of the output of the microphone is saturated, and the control sound including the high frequency that does not contribute to mute is produced by the speaker. There was a problem of being output from.

By adopting an A / D converter that does not saturate even when such loud noise is input, such a problem can be solved, but this leads to an increase in cost and a decrease in the silencing effect. The noise at the time of product purchase does not have the same periodicity as the noise of built-in equipment such as a compressor, and the noise reduction effect by feedback type active noise reduction cannot be expected. For this reason, it is not desirable to determine the performance of the A / D converter of the active silencer in consideration of the intensity of noise that is not the silencer.
JP-A-4-247497

  The conventional active silencer has a problem that when excessive noise that saturates the A / D converter is detected by a microphone, a control sound including a high-frequency component that does not contribute to the silence is output. Also, the feedback type active silencer cannot suppress non-periodic noise, and even when such noise is detected, an unnecessary control sound is output, and the active silencer becomes a new noise source. There was a problem of becoming.

  In particular, in a vending machine, a built-in device such as a compressor generates periodic noise, while large noise having no periodicity is generated at the time of product purchase or product replenishment. Also, noise outside the vending machine cannot be predicted. For this reason, when the periodic noise in the vending machine is suppressed by feedback-type active silencing, if excessive sound is detected by the microphone and the A / D converter is saturated, unnecessary control sound including high-frequency components is output. There was a problem that.

  The present invention has been made in view of the above circumstances, and an object thereof is to prevent unnecessary control sound from being output from a feedback-type active silencer by detecting excessive noise. In particular, an object of the present invention is to prevent a control sound including a high frequency that does not contribute to muffling from being output from a feedback-type active muffler by saturating an A / D converter that performs A / D conversion of a microphone output. .

  In addition, in a vending machine that suppresses periodic noise generated in the vending machine by feedback-type active noise reduction, when noise that is greater than the periodic noise that is the target of mute and has no periodicity occurs An object is to prevent unnecessary control sounds from being output. In particular, an object of the present invention is to prevent unnecessary control sounds from being output from the vending machine due to noise at the time of product purchase, product replenishment, or external noise.

  The feedback type active silencer according to the first aspect of the present invention generates a detection signal by A / D converting the output of the microphone, a speaker for outputting a control sound for suppressing the noise, a microphone for detecting the noise. An A / D converter, a reference signal generating means for obtaining a reference signal obtained by removing the control sound component from the detection signal, and a control sound for generating a control sound signal for driving the speaker based on the reference signal A feedback type active silencer comprising: a signal generation means; a noise level determination means for generating a determination signal by comparing the detection signal with a threshold value equal to or lower than a saturation level of the A / D converter; and based on the determination signal And control sound suppression means for suppressing the output of the control sound.

  In this feedback-type active silencer, a control sound signal for suppressing noise is generated based on a reference signal obtained by removing a control sound component from a detection signal after A / D conversion. At that time, the detection signal is compared with a threshold value equal to or lower than the saturation level of the A / D converter, and the output of the control sound is suppressed based on the comparison result. When the noise intensity level exceeds the saturation level of the A / D converter, a control sound including a high frequency that does not contribute to mute is output. By suppressing the output of such control sound, it is possible to prevent the feedback type active silencer from becoming a noise source when excessive noise is detected.

  Note that the control sound suppression means may be any means that suppresses output of control sound from the speaker. For example, it may be an attenuating means for attenuating the control sound signal, an attenuating means for attenuating the reference signal, or a means for controlling the control sound signal generating means to reduce the intensity level of the control sound. There may be. Further, the suppression of the output of the control sound by the control sound suppression means includes the interruption of the control sound.

  In addition to the above-described configuration, the feedback active silencer according to the second aspect of the present invention is configured such that the control sound suppression unit blocks the output of the control sound based on the determination signal, and the noise level determination unit detects the detection level. When the signal exceeds the first threshold value, the determination signal is output until a state where the detection signal is below the second threshold value equal to or lower than the first threshold value continues for a predetermined time.

  According to such a configuration, when the detection signal exceeds the first threshold, the output of the control sound is interrupted, and when the state below the second threshold continues for a predetermined time, the control sound is output. Output resumes. For this reason, when the noise which generate | occur | produces intermittently or the noise from which an intensity level fluctuates across a 1st threshold value is detected, it can prevent repeating on / off of a control sound. Since the speaker outputs an unnecessary sound when the output is interrupted or restarted, it is possible to prevent the unnecessary sound from being repeatedly output by configuring so that the control sound is not interrupted and restarted. The second threshold only needs to be equal to or less than the first threshold, and both may be different values or the same value.

  In addition to the above configuration, the feedback active silencer according to a third aspect of the present invention is configured such that the control sound signal generation unit uses the transfer coefficient storage unit that holds a transfer coefficient and the transfer coefficient to convert the reference signal to the above-described reference signal. A noise control filter for converting to a control sound signal and coefficient updating means for updating the transmission coefficient based on the detection signal are provided, and the transmission coefficient is initialized based on the determination signal.

  When an adaptive filter is employed in the control sound signal generation means and the transfer coefficient is updated based on the detection signal, undesired learning is performed when excessive noise occurs. That is, when the determination signal is output when the noise intensity level exceeds the first threshold value, the transmission coefficient is updated immediately before the determination signal is output, so that the transmission coefficient becomes an undesirable value. . For this reason, by initializing the transfer coefficient based on the determination signal, it is possible to prevent an undesirable control sound from being output when the silencing operation is resumed thereafter.

  A vending machine according to a fourth aspect of the present invention includes a temperature adjusting compressor and a feedback active silencer for suppressing noise of the compressor, and the feedback active silencer includes a microphone for detecting noise; A speaker that outputs a control sound for suppressing the noise, an A / D converter that generates a detection signal by A / D converting the output of the microphone, and a component of the control sound is removed from the detection signal In a vending machine having reference signal generation means for obtaining a reference signal and control sound signal generation means for generating a control sound signal for driving the speaker based on the reference signal, the detection signal is supplied to the compressor. Generates a judgment signal by comparing with a threshold value that exceeds the noise level and below the saturation level of the A / D converter. And noise level determining means that, based on the determination signal, constituted by a control of suppressing sound suppressing means the output of the control sound.

  With such a configuration, noise generated from the compressor can be effectively suppressed, while unnecessary control sound can be prevented from being output when excessive noise exceeding the saturation level of the A / D converter occurs. it can.

  In the vending machine according to the fifth aspect of the present invention, in addition to the above-described configuration, the compressor is housed in a housing, and the microphone is configured to detect noise outside the housing.

  A vending machine according to a sixth aspect of the present invention includes a temperature adjusting compressor and a feedback active silencer for suppressing noise of the compressor, and the feedback active silencer includes a microphone for detecting noise; A speaker that outputs a control sound for suppressing the noise, an A / D converter that generates a detection signal by A / D converting the output of the microphone, and a component of the control sound is removed from the detection signal Operation for detecting an artificial operation in a vending machine comprising reference signal generating means for obtaining a reference signal and control sound signal generating means for generating a control sound signal for driving the speaker based on the reference signal Detection means and control sound suppression means for suppressing the output of the control sound based on the detection result by the operation detection means. It is made.

  According to such a configuration, it is possible to suppress unnecessary control sounds from being output due to an artificial operation. That is, it is possible to suppress generation of control noise including a high frequency that does not contribute to silencing by generating excessive noise due to an artificial operation and saturating the A / D converter. For example, when excessive noise is generated by an operation of a product purchaser or an administrator who performs product replenishment, the generation of the noise can be predicted and output of unnecessary control sound can be suppressed.

  The vending machine according to the seventh aspect of the present invention is configured such that, in addition to the above configuration, the operation detection means detects an operation input for purchasing a product. According to such a configuration, since the control sound can be suppressed based on operations such as product selection and price insertion by the product purchaser, the output of unnecessary control sound can be suppressed early or reliably. Can do.

  In addition to the above configuration, the vending machine according to the eighth aspect of the present invention is configured such that the control sound suppression means suppresses the output of the control sound for a certain period of time based on the detection result by the operation detection means. The Since the generation period of noise generated by the purchaser's operation can be predicted, unnecessary control sound can be generated with a simple configuration by suppressing the output of the control sound for a certain period of time after the operation is detected. Output can be suppressed.

  A vending machine according to a ninth aspect of the present invention includes, in addition to the above configuration, a noise level determination unit that generates a determination signal by comparing the detection signal with a threshold value, and the control sound suppression unit includes the operation detection unit. The suppression of the control sound is started based on the detection result, and the suppression of the control sound is ended based on the determination signal. With such a configuration, it is possible to start suppression of the control sound when an artificial operation is detected, and end the suppression of the control sound based on the noise intensity level. For this reason, the suppression time of control sound can be varied according to the generation time of noise.

  In the vending machine according to a tenth aspect of the present invention, in addition to the above configuration, the operation detecting means detects an open state of the product replenishment opening / closing door, and the control sound suppressing means is configured to open the product replenishment opening / closing door. It is configured to suppress the output of the control sound during the state. Since it is difficult for an administrator of a vending machine to predict the work time for replenishment of products, etc., by controlling the control sound according to the opening period of the product replenishment opening and closing door, the simple configuration Unnecessary control sounds can be prevented from being output.

  The feedback type active silencer according to the present invention compares the noise detection signal with a threshold value equal to or lower than the saturation level of the A / D converter, and suppresses the output of the control sound based on the comparison result. For this reason, when excessive noise is detected, the A / D converter that performs A / D conversion of the output of the microphone is saturated, and the output of control sound including high frequency that does not contribute to muffling is suppressed. Can do.

  Further, the feedback type active silencer according to the present invention cuts off the output of the control sound when the detection signal exceeds the threshold value, and outputs the control sound when the state below the second threshold value continues for a predetermined time. Resumed. For this reason, it can suppress that the unnecessary sound output from a speaker accompanying interruption | blocking or resumption of an output is output repeatedly.

  Further, the vending machine according to the present invention suppresses the output of the control sound using the feedback type active silencer based on the detection result of the artificial operation. For this reason, output of unnecessary control sound can be suppressed with a simple configuration. Further, it is possible to detect the occurrence of excessive noise at an early stage or with certainty and suppress the output of the control sound.

Embodiment 1 FIG.
First, a vending machine equipped with a feedback-type active silencer according to Embodiment 1 of the present invention will be described below with reference to FIGS. The feedback type active silencer according to the present embodiment is used to silence noise generated in the vending machine.

<Vending machine>
1 (a) and 1 (b) are diagrams showing an example of a schematic configuration of a vending machine including a feedback-type active silencer according to Embodiment 1 of the present invention. FIG. 1A shows the vending machine 1 as viewed from the front, and FIG. 1B shows the state as viewed from the side. FIG. 2 is a view showing a state in which the open / close door 3 of the vending machine 1 of FIG. 1 is opened. FIG. 3 is a side sectional view of the vending machine 1 shown in FIG. The vending machine 1 is a vending machine that sells beverages enclosed in cans as products, and stores a large number of products in a casing 2 having a vertically long rectangular parallelepiped shape. Hereinafter, the left side in FIG. 1B will be described as the front and the right side as the rear.

  A front opening 2 </ b> A that opens the entire front surface is formed on the front surface of the housing 2, and the front opening 2 </ b> A is covered with an opening / closing door 3 that is attached to the housing 2 so as to be opened and closed. Legs 4 are attached to the bottom surface of the housing 2.

  In the center of the front surface of the open / close door 3, a money insertion area 5 is formed that is used when a user who wants to purchase goods inserts money. In this money insertion area 5, a coin insertion slot 5A for inserting coins, a bill insertion slot 5B for inserting banknotes, and a return lever 5C for instructing the return of the inserted coins are inserted. An input amount display portion 5D for displaying the total amount of money is provided. When the return lever 5C is operated after the coin is inserted, the bill is returned from the bill input port 5B, and the coin is returned from the coin return port 6 formed in the lower part of the open / close door 3.

  A display area 7 on which samples of merchandise are displayed is formed at the upper front of the open / close door 3. This display area 7 is provided with a purchase button 7A associated with each sample displayed, and after the user inserts money from the coin insertion slot 5A or the bill insertion slot 5B, By operating the purchase button 7A, a product corresponding to the purchase button 7A can be purchased. The product for which purchase is instructed is sent out to the lower part in the housing 2 and can be taken out from a horizontally long product outlet 8 formed in the lower part of the door 3.

  A money processing unit 9 is provided at a position corresponding to the back side of the money insertion area 5 on the front side of the central portion of the rear surface of the open / close door 3. The money handling unit 9 stores a money inserted from the coin insertion slot 5A and the bill insertion slot 5B, and performs a process for returning the money from the coin return slot 6 or the bill insertion slot 5B as necessary. It is. A control unit 10 is provided on the side of the money handling unit 9 in the center of the back of the open / close door 3, and the operation of the vending machine 1 is controlled by a processor provided in the control unit 10. ing.

  In the housing 2, a product storage portion 11 for storing products is formed in a region extending from the center to the top, and a cooling device 12 for cooling the products in the product storage portion 11 is provided in the housing 2. Located at the bottom inside. The product storage unit 11 is formed by arranging a cooling product storage unit 11 </ b> A and a heated product storage unit 11 </ b> B each having a vertically long shape side by side. A cooler 13 for cooling the air in the cooled product housing portion 11A is disposed below the cooled product housing portion 11A. On the other hand, the heater 14 for heating the air in the said heated goods accommodating part 11B is arrange | positioned under the heated goods accommodating part 11B. A product discharge port 15 for discharging a product from the product storage unit 11 to the product outlet 8 side is formed on the front side in the casing 2. A plurality of product discharge ports 15 are formed, and each product discharge port 15 is associated with the cooled product storage unit 11A or the heated product storage unit 11B.

  The cooling device 12 is for cooling the product accommodated in the cooled product accommodating portion 11A, and in addition to the cooler 13, the compressor 16 for compressing the refrigerant supplied to the cooler 13, the compressed A condenser 17 for liquefying the refrigerant and a cooling fan 18 for cooling the compressor 16 and the condenser 17 are provided.

  The cooler 13, the compressor 16, and the condenser 17 communicate with each other via piping, and cool the air in the cooled product storage portion 11 </ b> A via the cooler 13 by circulating the refrigerant through these portions. be able to.

  At the bottom of the open / close door 3, a horizontally long intake port 19 is formed in which both end portions are located in the vicinity of both left and right end portions of the open / close door 3. On the other hand, a horizontally long exhaust port 20 is formed at the bottom of the rear surface of the housing 2 so that at least a part thereof faces the intake port 19 in the front-rear direction. A partition plate 21 is provided below the product storage portion 11 in the housing 2, and a region below the partition plate 21 has an intake port 19 and an exhaust port 20 as openings, and the inside of the housing 2. A duct-like space 22 penetrating in the front-rear direction is formed.

  Among the components of the cooling device 12, a condenser 17, a cooling fan 18, and a compressor 16 are disposed in the duct-shaped space 22. On the other hand, the cooler 13 is disposed above the duct-shaped space 22.

  When the cooling device 12 is driven, the cooling fan 18 is rotationally driven, so that air outside the machine is sent into the machine through the air inlet 19. Then, the air sent into the machine passes around the condenser 17 and the compressor 16 in the duct-like space 22 and is sent out of the machine from the exhaust port 20, whereby the condenser 17 and the compressor 16 are cooled.

  Two openings 25 are formed symmetrically at a position below the opening / closing door 3 covering the front opening 2A of the housing 2 above the intake port 19 and below the product outlet 8. The two openings 25 are each composed of a plurality of slits extending in the left-right direction, and are formed by arranging the plurality of slits in the vertical direction so as to be parallel to each other. Each opening 25 is formed in a portion of the opening / closing door 3 that faces the front of the cooling device 12. Each opening 25 is provided with a speaker 23, and each speaker 23 is arranged facing forward.

  The two speakers 23 are arranged in a common internal space 26 formed in the open / close door 3. That is, the two openings 25 communicate with each other via the internal space 26 of the open / close door 3, and the speakers 23 are arranged so as to face the openings 25, respectively. A closed space that communicates with each other is formed.

  A microphone 24 for inputting noise generated in the vending machine 1 is provided on the front surface of the open / close door 3. The microphone 24 is disposed above the air inlet 19 on a center line that extends in the vertical direction through the central portion in the left-right direction of the door 3. As described above, the microphone 24 is arranged at the center in the left-right direction of the door 3 and the two speakers 23 are arranged symmetrically in the door 2 so that the distance between each speaker 23 and the microphone 24 is the same. be able to.

  Driving sound generated from the cooling device 12 in the housing 2 leaks out of the apparatus through the air inlet 19 and is input to the microphone 24 from the outside of the housing 2. In addition, sound output from the two speakers 23 leaks out of the apparatus through the opening 25 facing each other, and is input to the microphone 24 from the outside of the housing 2.

  In the present embodiment, an example of a vending machine in which the speaker 23 is disposed forward will be described. However, the speaker 23 can be disposed rearward, that is, toward the internal space 26.

  Next, a feedback-type active silencer according to Embodiment 1 of the present invention will be described below. This feedback-type active silencer is an active noise controller that silences a drive sound generated from the cooling device 12 in the vending machine 1 with a control sound output from the speaker 23.

<Active noise controller>
FIG. 4 is a block diagram illustrating a configuration example of a main part of the vending machine 1 of FIG. 1, in which a feedback-type active silencer 30 provided in the vending machine 1 is illustrated. In addition to the microphone 24 and the speaker 23, the active silencer 30 includes a microphone amplifier 31, an A / D converter 32, a DSP 33, a D / A converter 34, and a speaker amplifier 35. The microphone amplifier 31, the A / D converter 32, the DSP 33, the D / A converter 34, and the speaker amplifier 35 are provided in the control unit 10 of the open / close door 3, for example.

  The microphone 24 is input means for detecting noise. The output signal of the microphone 24 is amplified by the microphone amplifier 31 and then input to the A / D converter 32. The A / D converter 32 is an analog-digital conversion circuit that converts the amplified output signal of the microphone 24 into a digital signal and outputs the digital signal to the DSP 33 as a detection signal. In the A / D converter 32, the detection signal is sampled at a sampling rate sufficiently higher than the frequency of the noise to be silenced, for example, 3000 Hz.

  The speaker 23 is output means for outputting a control sound for canceling out noise by interference. The control sound signal generated by the DSP 33 is converted into an analog signal by the D / A converter and output to the speaker amplifier 35. The speaker 23 outputs a control sound based on the control sound signal amplified by the speaker amplifier 35.

  The DSP 33 is a mute processing circuit based on active control, and generates a control sound signal uj for canceling noise based on the detection signal ej from the microphone 24. Silencing by active control is a method in which sound waves are output from another sound source (secondary sound source) with respect to noise generated by a noise source, and noise is reduced by interference between sound waves. In the DSP 33, for example, a mute process is executed according to a predetermined control program.

<DSP>
FIG. 5 is a block diagram showing a configuration example of the DSP 33 of FIG. The DSP 33 includes a feedback filter 101, a reference signal generation unit 102, a noise control filter 103, a secondary filter 104, a coefficient update unit 105, a noise level determination unit 106, an output control unit 107, and a coefficient initialization unit 108.

  The feedback filter 101 is a filter processing unit that obtains a control sound component included in the detection signal ej from the control sound signal uj. The feedback filter 101 has characteristics until the control sound signal uj generated in the DSP 33 is fed back as the detection signal ej, that is, the control sound signal uj is emitted from the speaker 23 to the space as a control sound, and the microphone 24 And has a transfer coefficient close to the transfer characteristic until it is detected as the detection signal ej. By performing arithmetic processing using this transfer coefficient, the component of the control sound included in the detection signal ej is obtained from the control sound signal uj.

  The reference signal generation unit 102 is an arithmetic processing unit that obtains the reference signal xj by removing the component of the control sound included in the detection signal ej. Here, it is configured as a subtraction processing unit that subtracts the control sound component obtained by the feedback filter 101 from the detection signal ej detected by the microphone 24, and the reference signal xj obtained by removing the control sound component from the detection signal ej Is generated.

The noise control filter 103 is a filter processing unit that obtains a control sound signal uj from the reference signal xj. The noise control filter 103 includes a transfer coefficient storage unit 113 that holds a transfer coefficient W _j, and the reference signal xj from the reference signal generation unit 102 is converted into a control sound signal uj based on a predetermined transfer coefficient W _j. The In the noise control filter 103, the amplitude and phase of the reference signal Xj are adjusted and output as the control sound signal uj so that the noise is canceled by the control sound, that is, the detection signal ej is reduced.

  Since the detection signal ej is a signal including noise and control sound detected by the microphone 23, the DSP 33 for obtaining the control sound signal uj from the detection signal ej is a feedback-type active silencer. For this reason, if it is noise with periodicity, a favorable silencing effect can be acquired by using an appropriate transfer function. On the other hand, noise with no periodicity cannot be canceled out, and an unnecessary control sound that does not contribute to noise reduction may be output.

Here, it is assumed that the transfer coefficient of the feedback filter 101 is identified, for example, when the active silencer 30 is turned on and is fixed during the subsequent operation. In contrast, the noise control filter 103, an adaptive filter transfer coefficient W _j is automatically updated, transfer coefficient W _j is assumed to be updated at regular time intervals. The transfer coefficient is a coefficient for adjusting the amplitude and phase of the input signal. For example, the transfer coefficient may be a coefficient for adjusting the amplitude and phase for each frequency component.

  The secondary filter 104 is a filter for converting the noise component detected by the microphone 24 into a value at the position of the speaker 23, and is a filter processing unit corresponding to the feedback filter 101. Specifically, a filter having the same transfer function as the transfer characteristic of the feedback filter 101 can be used. The secondary filter 104 performs an operation of converting the reference signal xj from the reference signal generating unit 102 and outputting the converted signal to the coefficient updating unit 105.

The coefficient updating unit 105 performs an operation of updating the transfer coefficient W _j in the noise control filter 103 based on the detection signal ej and the reference signal Xj after conversion by the secondary filter 104. This update process is performed at regular time intervals, and the transfer coefficient W _j is optimized so that the sound pressure level of the detection signal ej becomes small.

Specifically, the transmission coefficient after updating with respect to the transmission coefficient W _j before updating is set as W _{j + 1} , and the transmission coefficient is updated based on the relational expression W _{j + 1} = W _j + μ × C (z) xj × ej. Μ in this relational expression is a constant called a step size parameter. In general, when the parameter μ is reduced, the time required until the transfer coefficient W _j is optimized becomes longer. _Assume that the rate at which the transfer coefficient W _j is updated is less than the sampling rate at which the A / D converter 32 A / D converts the detection signal.

  The noise level determination unit 106 compares the detection signal ej from the microphone 24 with a predetermined threshold value a1, and outputs the comparison result as a determination signal to the output control unit 107 and the coefficient initialization unit 108. The threshold value a1 is a value smaller than the intensity level at which the A / D converter 32 is saturated. When the intensity level of the detection signal ej, for example, the amplitude level exceeds the threshold value a1, a determination signal is output. The determination processing by the noise level determination unit 106 is performed at a sampling rate higher than that of the A / D converter 32, and a threshold value comparison is performed for each sampling data.

  The output control unit 107 blocks the output of the control sound signal uj based on the comparison result by the noise level determination unit 106. Specifically, when the intensity level of the detection signal ej becomes higher than the threshold value a1 and the noise level determination unit 106 outputs a determination signal, the output of the control sound signal uj is cut off, and the noise level determination unit 106 When the determination signal is not output from, the output of the control sound signal uj is resumed.

  Here, when the intensity level of the detection signal ej becomes higher than the first threshold value a1, the noise level determination unit 106 immediately outputs a determination signal and blocks the output of the control sound signal uj. On the other hand, during the output of the determination signal, the output of the determination signal is not terminated unless the state where the intensity level is equal to or lower than the second threshold value a2 continues for the predetermined determination time t1. That is, even if the intensity level of the detection signal ej falls below the second threshold value a2, the output of the determination signal is continued if it is only for a short time less than the determination time t1.

  The speaker 23 outputs a sound such as “buzz” when the input signal is interrupted or restarted. For this reason, when the output control unit 107 repeatedly turns on and off the control sound signal uj, the speaker 23 becomes a noise source. Therefore, when the noise level determination unit 106 determines using the determination time t1, if excessive noise is detected, the control sound signal uj is immediately interrupted, while intermittent noise or second noise is detected. When noise that fluctuates across the threshold value a2 is detected, the shut-off state is maintained, and the output control unit 107 is prevented from being repeatedly turned on and off.

  The second threshold value may be the same value as the threshold value a1. The determination time t1 is preferably determined in advance in consideration of, for example, the time required for the canned beverage to fall and stop in the vending machine 1.

  In the case of a vending machine that sells canned beverages and the like, a relatively small noise is generated from a temperature adjusting compressor and a cooling fan housed in the casing. Since such noise is periodic noise generated with the rotation operation, a good silencing effect can be obtained by feedback-type active silencing.

  On the other hand, the sound of falling canned beverages at the time of product purchase has no periodicity, and almost no silencing effect can be expected. In addition, since this falling sound is much louder than the noise of a compressor or the like, the noise intensity level saturates the A / D converter 32. When the noise intensity level exceeds the input range of the A / D converter 32, the noise suppression filter 103 generates a control sound signal uj including a high frequency component that does not contribute to mute, and the speaker 23 becomes a noise source.

  For this reason, the noise level determination unit 106 generates a determination signal based on the detection signal ej, and the output control unit 107 blocks the output of the control sound signal uj based on the determination signal, thereby not contributing to mute. It is possible to prevent unnecessary control sound including high frequency components from being output from the speaker 23.

The coefficient initialization unit 108 initializes the transfer function W _j in the noise control filter 103 based on the determination signal. When excessive noise is detected and a determination signal is output, the transfer coefficient W _j may be an undesired value by updating the transfer coefficient W _j immediately before the output of the determination signal. For this reason, by initializing the transfer coefficient based on the determination signal, it is possible to prevent an undesirable control sound from being output when the silencing operation is resumed thereafter. The transfer function W _j may be initialized every time a determination signal is output, and may be at the start of output of the determination signal or at the end of output.

  6 (a) and 6 (b) are diagrams showing an operation example of the active silencer 30 of FIG. FIG. 6A shows an example of the output signal of the microphone amplifier 31 with the horizontal axis as the time axis and the vertical axis as the signal level. FIG. 6B shows an active noise control operation with the horizontal axis as a time axis, that is, the on / off state of the output control unit 107. Here, a case where the threshold values a1 and a2 are equal will be described.

  When the amplitude level of the output signal of the microphone amplifier 31 exceeds a1, the output control unit 107 is switched from the on state to the off state. That is, an operation for reducing the output level of the control sound signal uj is performed. Here, the first threshold value a1 is smaller than the lower limit value b of the signal intensity at which the A / D converter 32 is saturated.

  The off state of the active noise control is continued until the determination time t1 elapses after the amplitude level of the output signal of the microphone amplifier 31 is at least a1 (= a2) or less. When the determination time t1 elapses, the active noise control is automatically turned on. Can be switched automatically. That is, after the amplitude level is higher than a1 and switched to the off state, when the signal level becomes a1 or less, and the signal level is a1 or less and the state continues for the determination time t1 or more, the on state is set. Can be switched.

  On the other hand, if the signal level again exceeds a1 (elapsed time t2 <t1) before the determination time t1 elapses after the signal level becomes equal to or less than a1, the output control unit 107 is not switched to the ON state. With this configuration, when the signal level is higher than a1 and the active noise control is turned off, the signal level is returned to the original on state when the state where the signal level is a1 or lower continues for the determination time t1 or longer. Therefore, it is possible to prevent the active noise control from being frequently turned on or off, and to suppress the restart of the output of the control sound when excessive noise is detected intermittently. Can do.

  According to the present embodiment, when it is detected from the comparison result between the intensity level of the detection signal ej and the threshold value a1 that the intensity level is higher than the threshold value a1, the output level of the control sound signal uj is automatically set. Therefore, when noise that saturates the A / D converter 32 is detected, a high frequency that does not contribute to mute can be prevented from being output as control sound. Further, since the output of the control sound signal uj is cut off based on the intensity level of the detection signal ej, an excessive control sound is output when excessive noise having no periodicity is input to the microphone 24. Can be suppressed.

  In the present embodiment, an example has been described in which the output control unit 107 suppresses output of unnecessary control sound by blocking the control sound signal uj output from the noise control filter 103. However, the present invention is not limited to such a case. For example, the reference signal xj may be blocked, or the output operation of the control sound signal uj by the noise control filter 103 may be controlled.

  Further, in the present embodiment, the example in which the output control unit 107 blocks the output of the control sound signal uj has been described, but the present invention is not limited to such a case. That is, it is only necessary to suppress the output of the control sound based on the determination signal, and is not limited to the one that completely blocks the output of the control sound. For example, the output control unit 107 may be an attenuation unit that attenuates the control sound signal uj and decreases the volume of the control sound.

Embodiment 2. FIG.
In the first embodiment, based on the comparison result by the noise level determination unit 106, the output control unit 107 blocks the output of the control sound signal uj, and the coefficient initialization unit 108 transfers the transfer coefficient w _j in the noise control filter 103. An example of initializing the above has been described. In contrast, in the present embodiment, a case will be described in which the output control unit 107 blocks the output of the control sound signal uj based on an operation input by the operator.

  FIG. 7 is a diagram showing an example of a schematic configuration of a vending machine provided with a feedback-type active silencer according to Embodiment 2 of the present invention, in which the opening / closing door 3 for replenishing products of the vending machine 1 is opened. The state is shown. In the vending machine 1, a DSP reset switch 42 is provided at the center of the back of the open / close door 3, and an open / close detection switch 41 is provided at the bottom of the back.

  The open / close detection switch 41 is a detection device for detecting opening / closing of the open / close door 3. The DSP reset switch 42 is an input device for initializing the DSP 33 when the vending machine 1 is installed. When the open / close door 3 is closed by operating the DSP reset switch 42, the initialization process of the DSP 33 is started after a predetermined time has elapsed since the open / close door 3 was closed.

In the initialization process of the DSP 33, for example, the transfer coefficient W _j held by the noise control filter 103 is initialized, and the operation of identifying the transfer coefficient of the feedback filter 101 is performed. Identification of the transfer coefficient of the feedback filter 101 is performed by outputting white noise from the speaker 23.

  FIG. 8 is a block diagram showing a configuration example of a main part of the vending machine 1 shown in FIG. 7, and shows another configuration example of the DSP 33. This DSP 33 is different from the case of FIG. 5 in that an operation detection unit 43 is provided.

  The operation detection unit 43 is a detection means for detecting an artificial operation input by a purchaser or an administrator, for example, opening / closing of the door 3, DSP reset switch 42, purchase button 7 </ b> A, operation of the return lever 5 </ b> C, coins The input of is detected.

  The output control unit 107 performs an operation of cutting off the output of the control sound signal uj based on the detection result by the operation detection unit 43. For example, when the manager opens the door 3 to replenish goods, the control sound signal uj is cut off by detecting that the door 3 has been opened. In addition, when the purchase button 7A is operated to purchase a product, the operation of the purchase button 7A is detected and the control sound signal uj is blocked. Further, the control sound signal uj is interrupted by detecting that a coin has been inserted into the coin insertion slot 5A for the purpose of purchasing a product.

  Generally, in the vending machine 1, when opening and closing the door 3, when a canned beverage falls in the vending machine 1 by operating the purchase button 7A, or when the return lever 5C is operated, the coin is returned from the coin return port 6. When it is done, a relatively loud sound is generated. When it is assumed that a loud sound is generated by such an artificial operation, the output of the control sound signal uj is blocked based on the detection result of the operation input.

The interruption of the control sound signal uj based on the output of the operation detection unit 43 predicts the occurrence of noise before it occurs. For this reason, even if the output control unit 107 interrupts the output of the control sound based on the output of the operation detection unit 43, if the coefficient update unit 105 stops updating the transmission coefficient W _j , It is not necessary for the coefficient initialization unit 108 to initialize the transfer coefficient W _j .

  When the operation detection unit 43 detects an artificial operation to interrupt the output of the control sound signal uj, the output control unit 107 resumes the output of the control sound signal uj after a certain period of time has elapsed. Since the generation time of noise due to human operation can be estimated in advance, the output of the control sound signal uj can be resumed after being interrupted for a certain time.

  However, regarding the opening / closing detection of the open / close door 3, when an opening operation or an unlocking operation is detected, the output of the control sound signal uj is immediately cut off, while a certain time has elapsed since the closing operation was detected. The output of the control sound signal uj may be resumed.

Embodiment 3 FIG.
In the second embodiment, a case has been described in which the output of the control sound signal uj is interrupted until a predetermined time has elapsed since the detection of the artificial operation. On the other hand, in the present embodiment, the control sound signal uj is started to be interrupted when the artificial operation is detected, but the output of the control sound signal uj is restarted based on the intensity level of the detection signal ej. To do.

  FIG. 9 is a block diagram showing a configuration example of a main part of the vending machine 1 according to the third embodiment of the present invention, and shows still another configuration example of the DSP 33. The DSP 33 is different from the case of FIG. 8 in that the output of the operation detection unit 43 is input to the noise level determination unit 106.

  In the noise level determination signal 106, the control sound signal uj cut-off operation and the subsequent restart operation when the detection signal ej exceeds the first threshold value a1 are performed in the same manner as in the first embodiment. On the other hand, when an artificial operation is performed, a determination signal is output based on the output of the operation detection unit 43. Therefore, the control sound signal uj is cut off in the same manner as in the second embodiment, but the subsequent restart operation is the same as in the first embodiment.

  That is, when an artificial operation is detected, the output of the control sound signal uj is immediately shut off. Thereafter, if the level of the detection signal ej does not exceed the first threshold value a1 within a certain time, the output of the control sound signal uj is resumed. The operation in this case is exactly the same as that in the second embodiment. However, if the detection signal ej exceeds the first threshold value a1 within a certain time, the output of the control sound signal uj is cut off until the state below the second threshold value continues for a predetermined determination time. With this configuration, when an artificial operation is detected, the output of the control sound is immediately interrupted, while the time during which the control sound is interrupted can be changed according to the generation time of the noise.

It is the figure which showed an example of schematic structure of the vending machine provided with the feedback type active silencer by Embodiment 1 of this invention. It is the figure which showed the state which opened the opening-and-closing door 3 of the vending machine 1 of FIG. It is a sectional side view of the vending machine 1 of FIG. FIG. 2 is a block diagram illustrating a configuration example of a main part of the vending machine 1 of FIG. 1, in which a feedback type active silencer 30 provided in the vending machine 1 is illustrated. FIG. 5 is a block diagram illustrating a configuration example of a DSP 33 in FIG. 4. It is the figure which showed the operation example of the active silencer 30 of FIG. It is the figure which showed an example of schematic structure of the vending machine provided with the feedback type active silencer by Embodiment 2 of this invention, and the state which opened the opening-and-closing door 3 for goods replenishment of the vending machine 1 is shown. ing. It is the block diagram which showed the structural example in the principal part of the vending machine 1 of FIG. 7, and the other structural example of DSP33 is shown. It is the block diagram which showed the structural example in the principal part of the vending machine 1 by Embodiment 3 of this invention, and the further another structural example of DSP33 is shown.

Explanation of symbols

1 Vending Machine 23 Speaker 24 Microphone 30 Active Silencer 31 Microphone Amplifier 32 A / D Converter 33 DSP
34 D / A converter 35 Speaker amplifier 41 Open / close detection switch 42 DSP reset switch 43 Operation detection unit 101 Feedback filter 102 Reference signal generation unit 103 Noise control filter 104 Secondary filter 105 Coefficient update unit 106 Noise level determination unit 107 Output control unit 108 Coefficient initialization unit

Claims (10)

A microphone for detecting noise, a speaker for outputting a control sound for suppressing the noise, an A / D converter for generating a detection signal by A / D converting the output of the microphone, and the control from the detection signal In a feedback type active silencer comprising: a reference signal generating means for obtaining a reference signal from which a sound component has been removed; and a control sound signal generating means for generating a control sound signal for driving the speaker based on the reference signal ,
Noise level determination means for generating a determination signal by comparing the detection signal with a threshold value equal to or lower than a saturation level of the A / D converter;
A feedback type active silencer comprising control sound suppression means for suppressing output of the control sound based on the determination signal. The control sound suppression means cuts off the output of the control sound based on the determination signal,
The noise level determination means outputs the determination signal until the state where the detection signal is below the second threshold value that is equal to or lower than the first threshold value continues for a predetermined time when the detection signal exceeds the first threshold value. The feedback-type active silencer according to claim 1. The control sound signal generating means includes a transfer coefficient storage means for holding a transfer coefficient, a noise control filter for converting the reference signal into the control sound signal using the transfer coefficient, and the transfer signal based on the detection signal. Coefficient updating means for updating the coefficient,
2. The feedback active silencer according to claim 1, wherein the transfer coefficient is initialized based on the determination signal. A compressor for temperature adjustment, and a feedback type active silencer for suppressing noise of the compressor,
The feedback active silencer includes a microphone that detects noise, a speaker that outputs a control sound for suppressing the noise, and an A / D converter that generates a detection signal by A / D converting the output of the microphone. Reference signal generation means for obtaining a reference signal obtained by removing the control sound component from the detection signal; and control sound signal generation means for generating a control sound signal for driving the speaker based on the reference signal. In vending machines with
Noise level determination means for generating a determination signal by comparing the detection signal with a threshold value exceeding the noise level of the compressor and less than or equal to the saturation level of the A / D converter;
A vending machine comprising control sound suppression means for suppressing the output of the control sound based on the determination signal. The compressor is housed in a housing,
The vending machine according to claim 4, wherein the microphone detects noise outside the casing. A compressor for temperature adjustment, and a feedback type active silencer for suppressing noise of the compressor,
The feedback active silencer includes a microphone that detects noise, a speaker that outputs a control sound for suppressing the noise, and an A / D converter that generates a detection signal by A / D converting the output of the microphone. Reference signal generation means for obtaining a reference signal obtained by removing the control sound component from the detection signal, and control sound signal generation means for generating a control sound signal for driving the speaker based on the reference signal. In the vending machine provided,
An operation detecting means for detecting an artificial operation;
A vending machine comprising control sound suppression means for suppressing output of the control sound based on a detection result by the operation detection means.   7. The vending machine according to claim 6, wherein the operation detecting means detects an operation input for purchasing a product.   8. The vending machine according to claim 7, wherein the control sound suppression means suppresses the output of the control sound for a predetermined time based on a detection result by the operation detection means. Noise level determination means for generating a determination signal by comparing the detection signal with a threshold value,
The said control sound suppression means starts suppression of the said control sound based on the detection result by the said operation detection means, and complete | finishes suppression of the said control sound based on the said determination signal. Vending machine. The operation detection means detects the open state of the product replenishment opening / closing door,
The vending machine according to claim 6, wherein the control sound suppression means suppresses the output of the control sound while the commodity replenishment opening / closing door is in an open state.

Images