JP2860026B2 - Vibration mitigation device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vibration mitigation device, and more particularly to a vibration mitigation device for reducing a vibration generated from a vibration source by applying a suppression vibration.

[0002]

2. Description of the Related Art Conventionally, an active noise control device has been known as a device for suppressing noise of a blower, an engine, and the like. The active noise control device can suppress a noise or a vibration by generating a sound wave or a vibration having a phase difference of 180 degrees with the noise or the vibration and having the same amplitude and causing interference. .

As prior art relating to periodic noise suppression, (1) “Adaptive Noise Canceling: Pr by WIDROW et al.
inciples and Application ”(PROCEEDI
NGS OF THE IEEE, VOL.63, NO.12, December, 1975) (2) Japanese Patent Publication No. Hei 3-501317 (3) Nakatsu et al., "Research on muffled sound active control technology focusing on vehicle interior sound field characteristics" (The Japan Society of Mechanical Engineers, No.910-52, Symposium Proceedings) and the like.

FIGS. 11 and 12 are a vector diagram and a vibration waveform diagram for explaining the principle of active noise control. Invention of the present application and the above prior arts (1) to (3)
Uses the principle for active noise control shown in FIG. 11 and FIG.

Referring to FIG. 11, a vector V0 indicates noise generated from a noise source. Vector V1 indicates the vibration of the noise source. Vector V2 has a phase 9 compared to vector V1.
Shows a vector delayed by 0 degrees. Vector V3 indicates an anti-phase additional sound to be added to suppress noise. The coefficients H0 and H1 are used to calculate the anti-phase additional sound vector V3, and the vectors V1 and V2
, Respectively.

FIG. 12 shows the relationship between the vectors V0 to V3 shown in FIG. 11 on the time axis. Referring to FIG. 12, the horizontal axis represents the passage of time in rotation angles (degrees), and the vertical axis represents the sound pressure level. Curve C0 indicates the noise, curve C1 indicates the vibration of the noise source,
A curve C2 indicates a vibration delayed by 90 degrees from the vibration C1 of the noise source, and a curve C3 indicates the opposite-phase additional sound.

In general, the noise vector V0 to be suppressed
Can not know. Therefore, in the principle of the active noise control, the vibration vector V1 of the noise source is detected, the vector V2 delayed by 90 degrees from the vector V1 is obtained, and the inverse to be added to suppress noise from these vectors V1 and V2. The phase-added sound vector V3 is calculated. Antiphase added sound vector V obtained by calculation
The opposite phase additional sound is emitted from the loudspeaker on the basis of No. 3, thereby reducing the noise level.

[0008]

In the active noise control device disclosed in the above prior art examples (1) and (2), it is necessary to obtain a vibration vector (ie, V2) delayed by 90 degrees of the vibration of the noise source. , Phased locked
A loop (hereinafter, referred to as "PLL") was required, and the circuit was complicated. Further, in the prior art examples (1) and (2), the time delay in the speaker drive path for generating the secondary additional sound is not considered.

On the other hand, in the prior art example (3), an additional digital filter is required to take into account the time delay in the speaker drive path for generating the secondary additional sound, and these are configured by software. The amount of calculation is large, and as a result, it is not suitable for high-speed processing.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and has as its object to provide a vibration mitigation device having a simple configuration and suitable for high-speed processing.

[0011]

SUMMARY OF THE INVENTION The vibration damping device according to the present invention, sampled at predetermined sampling time intervals vibrations emitted from the vibration source, and sampling means for outputting a sample vibration data U _n shifts at predetermined sampling time intervals vibration data U _n outputted from the sampling means, to n pieces of no vibration data U _n-1 delayed shift register means for outputting the U _0, vibration data U _n By applying the first and second vibration data and the first and second coefficient data having a phase difference of substantially 90 degrees among U ₀ to a first predetermined arithmetic expression, the vibration suppression signal is obtained. A first calculating means for outputting, a suppression vibration generating means for generating a suppression vibration for suppressing a vibration generated from the vibration source in response to the vibration suppression signal, and a vibration source. A residual vibration detecting means for detecting a residual vibration that is not suppressed by the suppressed vibration among the applied vibrations; and a storage means for storing propagation delay data of a signal generated in the suppressed vibration generating means and the residual vibration detecting means for each vibration frequency. A frequency detecting means for detecting the frequency of the vibration emitted from the vibration source; a propagation delay data in the storage means determined by the frequency detected by the frequency detecting means; and a residual vibration data detected by the residual vibration detecting means. By applying to two predetermined arithmetic expressions, the first and second arithmetic operations for the first arithmetic means are performed.
And second calculating means for outputting coefficient data of

[0012]

According to the vibration damping device of the present invention, approximately 90
The shift register means is used to obtain the first and second vibration data having a phase difference of degrees, so that the configuration can be simplified as compared with the case where a PLL is used.
In addition, propagation delay data generated in the suppressed vibration generating means and the residual vibration detecting means for each vibration frequency is stored in the storage means, and the propagation delay data is obtained by using the propagation delay data obtained by referring to the storage means. The first and second coefficient data for taking into account
Is obtained by a predetermined arithmetic expression. Therefore,
Compared with the conventional one which requires a large number of operations to take propagation delay into account, the amount of operations is reduced, and as a result, high-speed processing can be achieved.

[0013]

FIG. 2 is a block diagram of an active noise control system showing an embodiment of the present invention. The example shown in FIG. 2 shows an example in which the present invention is applied to suppress noise emitted from a blower. Referring to FIG. 2, a vibration sensor 12 for detecting vibration of rotating device 1 is attached to rotating device 1 that is a power source of the blower. Via the vibration sensor 12, the rotation frequency and the vibration vector of the rotating device 1 are detected. The detected vibration is provided to the A / D converter 9 via the filter 3 having an amplifier (not shown).

In the A / D converter 9, vibration is sampled at every predetermined sampling period, and vibration data is supplied to the adaptive digital filter 8. For example,
Rotating device 1 in active noise control system shown in FIG.
In the case where the vibration frequency is 250 to 500 Hz, assuming that the sampling time interval is 500 Hz, the number of samples is 20 at 250 Hz, and 5
There are 10 at 00 Hz.

A secondary additional sound signal is obtained by arithmetic processing in the adaptive digital filter 8, and the secondary additional sound signal is supplied to the speaker 6 via an amplifier (not shown) provided in the filter 5. Through the speaker 6, a secondary additional sound for suppressing noise generated from the rotating device 1 is emitted.

The error detecting microphone 4 detects the residual noise (or error) obtained by synthesizing the noise emitted from the rotating device 1 and the secondary additional sound, and supplies it to the filter 7. Filter 7
, The residual noise signal (or error signal) is supplied to the adaptive digital filter 8.

FIG. 1 is a block diagram of the adaptive digital filter 8 shown in FIG. Referring to FIG. 1, adaptive digital filter 8 includes n registers 80 to 8 _n-1.
A shift register constituted by n switches SW0 to SW _n-1 and a selector constituted by _n switches SW0 to SW _n-1 .
It includes coefficient multipliers 90 and 91, an adder 93, a coefficient calculator 94, and a delay data memory 95.

Each register 80 constituting the shift register
2 to 8 _n ^-1 have a time delay Z ^-1 corresponding to one sampling time interval in the A / D converter 9 shown in FIG. The number of registers corresponding to a time length of 90 degrees is
For example, 5 at 250 Hz, 2 or 3 at 500 Hz
Is equivalent to In the example shown in FIG. 1, it is assumed that a time delay of 90 degrees is obtained by two registers, and therefore, only the switch SWn _-2 is turned on. FIG.
Although the switches SW0 in the example shown SW _n-1 is used, in case the adaptive digital filter 8 shown in FIG. 1 is constituted by all software, switches SW0 through SW _n-1 is also software And functionally serves as a switch.

Switches SW0 to SW in selector
One of the _n-1 but by turning on, to no noise data U ₀ is delayed one of U _n-1 is given to the coefficient multiplier 90 or the coefficient calculating unit 94. Noise data U _n which is not delayed is supplied to the coefficient multiplier 91.

The coefficient multiplier 91 multiplies the vibration data U _n and the coefficient H1, provide the multiplied data to the adder 93. The coefficient multiplier 90 outputs the delayed vibration data U ₀ to U _{0
One of n-1} is multiplied by the coefficient H0, and the multiplied data is supplied to the adder 93. The adder 93 adds the two given multiplied data, and outputs the added data as a secondary additional sound signal Y.

The coefficient calculator 94 performs a calculation for updating the coefficients H0 and H1 used in the calculations in the coefficient multipliers 90 and 91. The delay data memory 95 has 2
Data on the propagation delay of the signal in the next additional sound generation path is stored for each vibration frequency. Delay data memory 9
The data stored in 5 is used in a calculation process for updating a coefficient in the coefficient calculation unit 94.

Basic operations for noise suppression control are performed using a shift register, a selector, coefficient multipliers 90 and 91, and an adder 93. That is, _assuming that the noise data delayed by 90 degrees with respect to the current noise data Un is _Un-2 , the switch SW shown in FIG.
_n-2 turns on, and therefore the coefficient multipliers 90 and 91
Are provided with the vibration data _Un-2 delayed by 90 degrees and the current vibration data _Un , respectively. Therefore, the calculation by the following equation is performed by coefficient multipliers 90 and 91 and adder 93, and output signal Y of adder 93 is output as a secondary additional sound signal.

Y = _Un * H1 + Un _-2 * H0 (1) As a result, an anti-phase additional sound vector V3 is obtained as a secondary additional sound signal Y based on the principle diagram shown in FIG.
Based on the secondary additional sound signal Y, a secondary additional sound for suppressing noise is emitted from the speaker 6.

In the coefficient calculation section 94 shown in FIG. 1, the calculation processing expressed by the following equation is performed for updating the coefficient.

H1 = H1′-2 * α * U _nm * E (2) H0 = H0′-2 * α * U _n-2-m * E (3) In (3), the coefficients H1 and H0
Indicates an updated coefficient, and coefficients H1 'and H0' indicate old coefficients before updating. Α is a small positive value (for example, 0.01), and E corresponds to error data (residual noise data) detected by the sensor 4 shown in FIG.

The data m is stored in the delay data memory 95 shown in FIG. 1 for each vibration frequency. Since the delay time length of the signal in the secondary additional sound generation path is changed according to the frequency of the noise, the delay data memory 95
, Data m is read according to the noise frequency.

[0027] In the above equation (1), with respect to input vibration data U _n, it is assumed that the vibration data delayed 90 degrees is output from the register 8 _n-2 shown in FIG. In other words, it is assumed that a 90 degree delay is obtained through two registers. Α is determined by the stability or responsiveness of the system required in the noise control system.

FIG. 3 shows the relationship between the delay time difference between the two vibration data used in the adaptive digital filter shown in FIG. 1 and the number of updates of the coefficients required for the error (residual noise) to stably converge. It is a graph. Referring to FIG. 3, the horizontal axis indicates a delay time difference (degree) between two pieces of vibration data, and the vertical axis indicates the number of updates of a coefficient required for stable convergence. As can be seen from FIG. 3, it is most preferable that the delay time difference is 90 degrees in order to stably converge the coefficients in a short time, but it is not always necessary to fix the delay time to 90 degrees. Within the range, the time required for convergence does not change much. Therefore, it is sufficient that the delay time difference between the current vibration data U _n and its 90-degree delay data (U _{n-2 in} the example shown in FIG. 1) is set to approximately 90 degrees.

FIG. 4 is a flow chart showing a calculation processing flow in the adaptive digital filter shown in FIG. Referring to FIG. 4, in step 21, the vibration data is shifted by the shift register shown in FIG.
Therefore, after one sampling time interval has elapsed, the current vibration data _Un is output as delayed vibration data _Un-1 . At the same time, new vibration data _Un is input to the first register _8n-1 .

In step 22, the delay data memory 95 is referred to. That is, the delay data memory 95 is referred to using the current vibration frequency, and the data m is read.

In step 23, new coefficients H1 and H0 are obtained by calculation by applying the data m to the above equations (2) and (3).

In step 24, new coefficients H1,
By applying H0 to the above equation (1), secondary additional sound data Y is obtained by calculation.

In step 25, a secondary additional sound is generated from the speaker 6 based on the obtained secondary additional sound data Y.

The post-processing of step 25 returns to step 21, and the processing of steps 21 to 25 is repeated. As a result, both coefficients H1 and H0 converge to 0,
Control for noise suppression is performed.

FIG. 5 is a graph showing the progress of updating the coefficients H1 and H0 updated by the equations (2) and (3). Referring to FIG. 5, the horizontal axis indicates a change in coefficient H1, and the vertical axis indicates a change in coefficient H0. As can be seen from FIG. 5, both the coefficients H1 and H0 converge quickly.

FIG. 6 is a graph showing a change in the level of the error data E corresponding to the update of the coefficients H1 and H0 shown in FIG. Referring to FIG. 6, the horizontal axis represents the number of updates of coefficients H1 and H0, and the vertical axis represents error E (decibel).
In FIG. 6, the error E before updating the coefficient is set to 0 dB. As can be seen from FIG. 6, the error E is stably converging.

FIG. 7 is a graph showing the progress of the updating of the coefficients H1 and H0 in the case where the equation (2) and the data m in the addition are not considered. In the example shown in FIG. 7, the delay time difference is also shifted by one register shown in FIG.
1, H0 changes unstablely and it takes a long time to converge.

FIG. 8 is a graph showing a change in the level of the error data E corresponding to the update of the coefficients H1 and H0 shown in FIG. As can be seen by comparing FIG. 8 with FIG.
May be temporarily large and therefore unstable,
It can be seen that the time required for convergence is long.

In the following description, data m to be stored in the delay data memory 95 will be described. In the secondary additional sound generation path, as described above, a propagation delay that changes according to the vibration frequency occurs. Therefore, the propagation delay is measured in advance, and the propagation delay data is stored in the delay data memory 95.

More specifically, in the measurement mode, a sine wave signal is output from the adaptive digital filter 8 shown in FIG. 2 as a signal Y, and the filter 5, speaker 6, microphone 4, filter 7 shown in FIG. To measure the delay time of the signal in the path returning to the adaptive digital filter 8 via the. The delay that occurs in this path includes the air propagation delay from the speaker 6 to the microphone 4. Since the filters 5 and 7 include an amplifier (not shown), their delay is also included in the measurement. The delay time in the above path is measured for each vibration frequency.

For example, assuming that the sampling frequency of vibration data U is 5000 Hz, the delay time length of one register shown in FIG. 1 is 200 μsec. Therefore, when a 500 Hz sine wave signal is generated as the output signal Y and the time delay in the above path is measured, the measurement result shown in FIG. 9 is obtained.

Referring to FIG. 9, the horizontal axis indicates the frequency (kHz) of the sine wave signal, and the vertical axis indicates the delay time (msec). The delay time is defined between the signal Y (sine wave signal) and the error signal E as shown in FIG.

Therefore, when the data m to be used in the equations (2) and (3) is obtained in consideration of the delay time for each frequency, the data shown in the following Table 1 is obtained as an example. This data is stored in the delay data memory 95 shown in FIG.

[0044]

[Table 1]

Therefore, since the data shown in Table 1 is stored in the delay data memory 95, when the coefficients H1 and H0 are calculated using the equations (2) and (3), the data is calculated based on the vibration frequency. Data m is read, and an operation is performed using read data m.

As described above, in the active noise control system shown in FIG. 2, the adaptive digital filter 8 shown in FIG.
Is used, vibration data having a time delay of 90 degrees can be obtained through a simple configuration, that is, via a shift register. In addition, the propagation delay of the signal in the secondary additional sound generation and residual noise detection paths can be easily considered using equations (2) and (3). Active noise control system is obtained.

In the above embodiment, an example is shown in which the present invention is applied to an active noise control system. However, it is pointed out that the present invention can be widely applied to a vibration mitigation device for suppressing vibration generated from a vibration source. Is done.

[0048]

As described above, according to the present invention, a shift register having a phase difference of about 90 degrees by using a shift register means.
Vibration data can be obtained and the propagation delay of the signal generated in the suppressed vibration generating means and the residual vibration detecting means can be considered using a simple arithmetic expression. The device was obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram of an adaptive digital filter used in the active noise control system shown in FIG.

FIG. 2 is a block diagram of an active noise control system showing one embodiment of the present invention.

FIG. 3 is a graph showing a relationship between a delay time difference between two vibration data used in the adaptive digital filter shown in FIG. 1 and the number of updates of a coefficient required for stable convergence of an error.

FIG. 4 is a flowchart of a calculation process in the adaptive digital filter shown in FIG. 1;

FIG. 5 is a graph showing the progress of updating coefficients H1 and H0 updated by arithmetic expressions (2) and (3).

FIG. 6 is a graph showing a change in an error E corresponding to updating of the coefficient shown in FIG. 5;

FIG. 7 is a graph showing the progress of updating coefficients H1 and H0 when data m in arithmetic expressions (2) and (3) is not considered.

8 is a graph showing a change in an error E corresponding to updating of the coefficient shown in FIG. 7;

FIG. 9 is a graph showing measurement results between a vibration frequency and a time delay generated in a secondary additional sound generation and residual noise detection path.

FIG. 10 is a graph defining a delay time shown in FIG. 9;

FIG. 11 is a vector diagram for explaining the principle of active noise control.

FIG. 12 is a vibration waveform diagram for explaining the principle of active noise control.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Rotary apparatus 2 Vibration sensor 3, 5, 7 Filter 4 Error detection microphone 6 Speaker 8 Adaptive digital filter 9 A / D converter 90, 91 Coefficient multiplier 93 Adder 94 Coefficient operation part 95 Delay data memory 80 to 8 _n-1 registers SW0 through SW _n-1 switch U, U _n to U ₀ vibration data E error signal Y 2 order additional sound signal

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-4-109124 (JP, A) JP-A-3-284098 (JP, A) JP-A-3-149999 (JP, A) 172053 (JP, U) Table 3-3-1317 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G10K 11/00

Claims (1)

(57) [Claims] 1. A vibration damping apparatus to mitigate vibration generated from the vibration source, sampled at predetermined sampling time intervals vibrations, sampling means for outputting a sample vibration data U _n, the sampling the vibration data U _n outputted from the means shifted at the predetermined sampling time intervals,
It n pieces of no vibration data U _n-1 delayed shift register means for outputting the U _0, the first and second vibration data having a phase difference of substantially 90 degrees out of vibration data U _n to U ₀ and A first calculating means for outputting a vibration suppression signal by applying the first and second coefficient data to a first predetermined calculation expression; and a signal generated from the vibration source in response to the vibration suppression signal. Suppressing vibration generating means for generating a suppressing vibration for suppressing the vibration to be applied; residual vibration detecting means for detecting a residual vibration that is not suppressed by the suppressing vibration among the vibrations generated from the vibration source; Storage means for storing propagation delay data of a signal generated in the suppression vibration generation means and the residual vibration detection means; Detecting means for applying the propagation delay data in the storage means determined by the frequency detected by the frequency detecting means and the residual vibration data detected by the residual vibration detecting means to a second predetermined arithmetic expression By doing so, the first
And a second calculating means for outputting the first and second coefficient data for the calculating means.