JPH0553589A - Active noise controller - Google Patents

Abstract

PURPOSE:To effectively control noises originating plural noise sources which do not correlate with one another with simple constitution. CONSTITUTION:This active noise controller is equipped with a control sound source 3 which generates a control sound to interfere with a noise and reduces the noise at an evaluation point, a means 5 which detects the residual noise at a specific position after the interference, a means 50 which detects signals regarding the noise generation states of plural noise sources having no mutual correlation and generates a reference signal, and a control means 7 which outputs a signal for driving the control sound source 3 according to the output signal of the residual noise detecting means 5 and the output signal of the reference signal generating means 5; and the reference signal generating means 50 adds one kind of random signal to at least one stationary determination signal to generate the reference signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an active noise control system for actively reducing noise in a passenger compartment of an automobile or a passenger compartment of an aircraft.

[0002]

2. Description of the Related Art As a conventional active noise control device of this type, there is, for example, one shown in FIG. 4 of British Patent Publication No. 2149614.

This conventional device is applied to a cabin of an aircraft or a closed space similar to this, and is provided with loudspeakers 103a, 103b, 103c and microphones 105a, 105b, 105c, 105d in the closed space 101, and Control sounds that interfere with noise are generated by the speakers 103a, 103b, 103c, and the residual signal (residual noise) is measured by the microphones 105s, 105b, 105c, 105d. These loudspeakers 103a, 103b, 103c and microphones 105a, 105b, 105c, 105d are connected to a signal processor 107, and the signal processor 107 measures the fundamental frequency of the noise source measured by the fundamental frequency measuring means and the microphones 105a, 105b. , 105c, 105
and the loudspeakers 103a and 1a so as to minimize the sound pressure level in the closed space 101.
A noise signal is output to 03b and 103c.

Here, three loudspeakers 103a, 103b, 103c and four microphones 105a, 105b, 105c, 105d are provided in the closed space 101, but for simplification of description, each is 013.
It is assumed that each of a and 105a is provided. Now, let the transfer function from the noise source to the microphone 105a be H, and let the loudspeaker 103a to the microphone 105a
Assuming that the transfer function up to is C and the sound source information signal generated by the noise source is X _p , the residual signal E observed by the microphone 105a is E = X _p · H + X _p · G · C. Here, G is a transfer function required for muffling. When the noise is completely canceled at the noise reduction target point, E = 0. At this time, G is G = -H / C. Then, G that minimizes the microphone detection signal E is obtained, and the filter coefficient in the signal processor 107 is adaptively updated based on this G. As a means for obtaining the filter coefficient so as to minimize the microphone detection signal E, the LMS algorithm (Le
ast Mean Square).

Further, as shown in FIG. 4, when a plurality of microphones are installed, the respective microphones 105a, 10a
The signals are controlled so that the sum total of the signals detected by 5b, 105c, and 105d is minimized.

[0006]

However, since the above-mentioned conventional device is configured to reduce the noise generated from a single noise source, for example, an engine, when the noise is transmitted from a plurality of noise sources, for example, However, it has been unsuitable for reducing the noise transmitted from a plurality of noise sources, such as having to apply the noise reduction target to only one of the noise sources.

On the other hand, a configuration in which a plurality of the above-mentioned conventional devices are mounted in parallel for a plurality of noise sources and each of them is independently driven is conceivable, but this causes an increase in size, complexity, and cost of the device. Not realistic.

Further, in the vehicle, there are various kinds of noise such as road noise transmitted from the suspension in addition to vehicle interior noise due to engine rotation. Therefore, noise generated from the engine as a single noise source is reduced. With the configuration,
There was a risk that appropriate noise reduction could not be achieved. In other words, the noise generated from the engine is targeted for the steady definite signal consisting of only the sine component, while the wind noise,
Road noise, exhaust noise, and the like are often of random nature, and there was a problem that all could not be silenced reliably by using only a steady definite signal as a reference signal.

Therefore, an object of the present invention is to provide an active noise control device capable of effectively reducing noises from a plurality of noise sources on the basis of a steady definite signal and a random signal with a simple structure. ..

[0010]

In order to solve the above-mentioned problems, the present invention provides a control sound source for reducing a noise at an evaluation point by generating a control sound that interferes with noise, and a residual of a predetermined position after the interference. Means for detecting noise; means for generating a reference signal by detecting signals relating to the noise generation state of a plurality of noise sources that are not correlated with each other; output signal of the residual noise detecting means and output of the reference signal generating means And a control means for outputting a signal for driving the control sound source based on a signal, wherein the reference signal generating means is
It is characterized in that one kind of random signal is added to at least one steady definite signal to make it a reference signal.

[0011]

The control means outputs a signal for driving the control sound source based on the output signal of the residual noise detecting means and the output signal of the reference signal generating means. As a result, the control sound source can generate a control sound that interferes with the noise to reduce the noise at the evaluation point. In this case, the reference signal generating means has a configuration in which one kind of random signal is added to at least one steady definite signal to make it a reference signal. Therefore, the noise relating to the steady definite signal and the noise relating to the random signal are combined into one reference signal. Can be reduced at the same time.

[0012]

Embodiments of the present invention will be described below. In addition,
The description will be given using the vehicle interior space as an example.

FIG. 1 is a block diagram of an active noise control system according to an embodiment of the present invention, in which a loudspeaker 3 as a control sound source and a residual noise detecting means are provided in a vehicle compartment 1 which is a closed space. A microphone 5 is provided, and each is connected to a controller 7 as control means.

Although only one loudspeaker 3 and one microphone 5 are provided for simplicity of explanation, for example, a plurality of them are provided, and the loudspeakers are left and right corresponding to the front and rear seats of the vehicle. The microphones may be arranged at the doors, and the microphones may be provided at the headrest positions of the seats.

The controller 7 comprises an adaptive digital filter 9 and an adaptive controller 11, the microphone 5 is connected to the adaptive controller 11, and the loudspeaker 3 is connected to the adaptive digital filter 9 via an amplifier 13. It is connected. The filter coefficient of the adaptive digital filter 9 is adapted to be adaptively rewritten by the adaptive controller 11. Note that the controller 7 performs a series of operations by digital operation, but the description of the inevitable elements such as an A / D converter and a D / A converter is omitted.

The noise in the passenger compartment 1 is caused by, for example, the engine 13,
Power plant 17 consisting of transmission 15
, The transverse link 19, the upper link 21,
The noise (primary sound) is transmitted from the suspension 27 composed of the third link 23 and the sub-frame 25, and a signal related to the noise generation state of these noise sources is detected and added by the adder 29 to obtain at least one reference value. The signal is input to the adaptive digital filter 9 as a signal.

Of the signals input to the adder 29,
The steady definite signal x ₁ consisting of only the sine component is the engine 13
Based on the detection signal of the rotation signal detector 31. That is, the rotation pulse obtained from the rotation signal detector 31 is usually a signal of one pulse for every 180 degrees of the crankshaft rotation angle. This detection signal is converted into one or a plurality of sine waves having a cycle of n times or 1 / n (n is an integer) of the pulse interval based on the pulse train by the waveform calculator 33, output after gain adjustment, The confirmation signal x ₁ is used.

The other steady state determination signal x ₂ is based on the detection signal of the first vibration sensor 35. That is, the first vibration sensor 35 is composed of a piezo element or the like, is attached to the surface of the power plant 17, and detects the vibration at the attachment point. Since the vibration of the power plant 17 is caused by the rotational movement of the engine, explosion, rotation of the gear system, etc., the main components are the frequencies of these rotational speeds and a complex sine wave composed of harmonics thereof. Therefore, the engine and gear noise can also be controlled by using the detection signal of the first vibration sensor 35 as the reference signal. However, the output of the first vibration sensor 35 is gain-adjusted by the equalizer 37 so that each sine component is reliably controlled. That is, when the output signal of the first vibration sensor 35 is used, there is a possibility that the amplitude difference between the sine wave components becomes excessively large, and in that case, the weak component cannot be controlled. Therefore, the equalizer 37 avoids this. is doing. In this case, the equalizer 37 is required, but the noise caused by both the engine and the transmission is the same as the reference signal x.
There is a merit that it can be controlled by ₂ .

In this embodiment, both x ₁ and x ₂ are input to the adder 29 as the steady decision signal, but only one of them may be input.

The random signal is based on the detection signal of the second vibration sensor 39 provided in the sub-frame 25 at the substantially central position of the suspension 27, which is targeted for road noise, for example. The second vibration sensor 39 is
The vibration signal, which is the basis of the road noise transmitted from the left and right wheels, is detected and output. This second vibration sensor 39
The detection signal of is subjected to gain adjustment by an amplifier (or equalizer) 41 and then input to the adder 29 as a random signal x ₃ .

The random signal x ₄ may be input to the adder 29 instead of the random signal x ₃ . That is, the third vibration sensors 43 and 45 may be provided at symmetrical positions on the left and right of the suspension 27, and the output may be output as a random signal pair x ₄ through an amplifier (or equalizer) 47. In this case, subframe 25
Although the number of sensors is larger than the case where one second vibration sensor 39 is provided at the center of the above, there are the following advantages.

That is, in the case of noise having a random property such as road noise, it is necessary to perform so-called feedforward control in which the noise is predicted from the vibration signal detected by the sensor and the noise is canceled. In other words, after detecting the signal, it is necessary to have sufficient time to release the control sound (secondary sound), and the more time is available, the more accurate calculations can be performed and the processing required by the cheaper computing unit can be performed. Can be performed. Therefore, when the plurality of sensors 43 and 45 are provided near the spindle upstream of the vibration transmission path, the transmission time of the primary sound from the detection point to the vehicle interior is delayed by the distance from the vehicle interior, and the margin is increased accordingly. It is advantageous because it gives time.

As described above, the steady definite signal x ₁ or x ₂ , or x ₁ and x ₂ and the random signal x ₃ or x ₄ are added by the adder 29, and the adaptive digital filter 9 is used as one reference signal x. It is configured to be input to.

Therefore, the rotation signal detector 31, the waveform calculator 33, the first vibration sensor 35, the equalizer 37, the second
Vibration sensor 39, equalizer 41, third vibration sensor 4
3, 45, the equalizer 47, and the adder 29 constitute the reference signal generating means 50 in this embodiment, and add one kind of random signal to at least one steady definite signal,
It is characterized by using at least one reference signal.

Next, the operation will be described.

The active noise control system of this embodiment generates the steady-state confirmation signals x ₁ and x ₂ based on the detections of the rotation signal detector 31 and the first vibration sensor 35, and the second vibration sensor 39 or the second vibration sensor 39. The random signal x ₃ or x ₄ is generated based on the detection of the three vibration sensors 43 and 45, and the adder 29
To the controller 7 as one reference signal x.

The reference signal x is filtered by the adaptive digital filter 9 of the controller 7 and output as the drive signal y, which is supplied to the loudspeaker 3 via the amplifier 13. The control sound (secondary sound) emitted from the loudspeaker 3 into the vehicle interior 1 has an opposite phase to the noise transmitted from the power plant 17 and the suspension 27 into the vehicle interior 1, and the residual sound is left as a result. The pressure component is detected as residual noise by the microphone 5, and is input to the adaptive controller 11 of the controller 7 as a noise signal e.

The adaptive controller 11 outputs a signal s for updating the filter coefficient of the adaptive digital filter 9 so that the sum of squares of the noise signal e is minimized. The calculation performed by the adaptive controller 11 uses, for example, the LMS algorithm which is a kind of steepest descent method, and a general description of the control method will be given later.

In this way, by making the reference signal x the sum of a plurality of steady decision signals x ₁ and x ₂ and one type of random signal x ₃ or x ₄ , it is possible to reduce the widest range of types of noise. ..

This will be described below.

The spectrum of the steady definite signal is a line spectrum as shown in FIG. 2, and the random signal has a spectrum continuously distributed with respect to frequency. Therefore, the discrete fixed signal is almost 1 at the frequency points.
While the energy of the 00% steady-state fixed signal, it is only the component of the random signal at other frequencies.
Therefore, the overlap of both signal components on the frequency axis is substantially zero. Therefore, all these signals can be controlled as long as the adaptive digital filter 9 has a sufficient degree of freedom.

As the adaptive digital filter 9, for example, in the case of a finite impulse response type filter, the following tap number (filter length) may be used. That is, the number of taps is larger than the larger one of the following A and B.

A: The number of taps at which the filter length is equal to or longer than the impulse response length of the vibration signal to the sound pressure signal.

B: Several times (four times or more) the number of steady definite signals included in the reference signal.

A means a filter length for sufficiently controlling a random signal. Here, the impulse response length is the time until the amplitude becomes 0 when a pulsed vibration input is input, and the filter length (the number of taps) is the impulse response length divided by the sampling period.

B indicates the filter length that can control the line spectrum of each steady determinate signal and the random signal component between them, point by point. One line spectrum (steady fixed signal)
Can control the phase and amplitude with two taps. Therefore, 4n taps (n: the number of steady-state fixed signals) are required when controlling all the steady-state fixed signals and the random signals between them, one point at a time.

It should be noted that the random signal component added as the reference signal is one kind, but if other random signal components are present, the random signal components will interfere with each other in all frequency bands, and each of them will be treated by one adaptation. It becomes impossible to control with the digital filter 9. Therefore, it can be seen that the above-described configuration in which the random signal component is one type enables noise reduction in the maximum range.

Another explanation of this point will be given with reference to FIG.

That is, consider the case where the vibrations of the front and rear suspensions are detected as random signal components as shown in FIG. 3 and are simultaneously input to one adaptive digital filter. The time required for these noises to reach the evaluation points 111a and 111b for reducing the noise of the front and rear seats in the vehicle compartment is different and the timings are different. In FIG. 3, A ₁ , A ₂ , B ₁ and B ₂ are the propagation times of the route,
At the evaluation point 111a in the front seat, A ₁ <B _1, and the noise due to the front suspension reaches quickly,
At the rear seat evaluation point 111b, A ₂ > B ₂ . Therefore, if the vibration detection signal is input to the adaptive digital filter at the same timing, it is not possible not only to efficiently and efficiently attenuate the noise due to the front and rear suspensions that have different arrival times, but it is also difficult to properly control one of them. There is a risk that Therefore, it is necessary to use one type of random signal component to be detected, which is the most effective.

The suspension 2 shown in FIG.
In the example in which the third vibration sensors 43 and 45 are provided at two positions in FIG. 7, the positions are symmetrical to each other, so that the noise arrival timings in the passenger compartment 1 are almost the same, and there is no timing deviation from the vibration detection signal. Does not occur. Further, in the example in which the second vibration sensor 39 is provided in the sub-frame 25, since it is provided in the center of the sub-frame 39, the vibrations from both the left and right wheels can be detected at the same time, and the noise arrival timing and the vibration in the vehicle compartment 1 are similarly detected. There is no deviation from the detection signal, and the above problem does not occur.

Further, in this embodiment, the waveform calculator 33, the equalizer 37, the amplifier 41 or the amplifier 47 is provided, but the respective gains are as follows. For example, if the amplitude of the steady-state deterministic signal is larger than that of the random signal, the amplitudes of the portions other than the peaks and valleys of the waveform are always smaller than that. You are only using the dynamic range. Especially when the phase of each steady definite signal is the same,
The valleys overlap each other to further increase the amplitude, and the dynamic range of the adaptive digital filter 9 is determined by the peaks and valleys of the sine wave of the steady decision signal, resulting in a loss of control performance. Therefore, the amplitude (eg, effective value) of the steady decision signals x ₁ and x ₂ is set to the random signal x ₃
Or it should be equal to or less than x ₄ . Alternatively, the absolute value of these amplitudes is made constant by a regulator (not shown). By doing so, the effective amplitude value of the reference signal x input to the adaptive digital filter 9 via the adder 29 can be made substantially constant, and the dynamic range of the adaptive digital filter 9 can always be effectively utilized. The maximum noise reduction capability can be demonstrated.

Next, the control method will be generally described.

Now, assuming that a plurality of microphones and loudspeakers are provided, the noise signal detected by the l-th microphone is e _l (n), and the l-th microphone when there is no control sound from the loudspeaker. The noise signal detected by e _pl (n), the j-th transfer function FIR (finite impulse response) function between the m-th loudspeaker and the l-th evaluation point (j = 0, 1, 2) ,…, I _c −
When the term 1) is represented by a digital filter, the filter coefficient is C _lmj , and the reference signal, that is, the sound source information signal is x.
_p (n) and the reference signal x _p (n) are input, and the i-th (i = 0, 1, 2, ..., I _k -1) coefficient of the adaptive digital filter for driving the m-th loudspeaker is set. W _mi

[0044]

[Equation 1]

Is satisfied. Here, the term with (n) is
Both are sample values at sampling time n, L is the number of microphones, M is the number of loudspeakers,
I _c is the tap number (filter order) of the transfer function C _lm expressed by the FIR digital filter, and I _k is the tap number (filter order) of the adaptive digital filter.

In the above equation (1), "Σ W _mi x
_{The term “p} (n−j−i)” (= y _m ) represents the output when the signal x _p is input to the filter (coefficient W _m ) for each output channel in the adaptive digital filter, and “Σ
The term “C _lmj · {Σ W _mi · (n−j−i)}” means that the signal energy input to the mth speaker is output as acoustic energy from the speaker and passes through the transfer function C _lm in the vehicle interior. It represents the signal upon reaching the l th microphone, further _{"Σ Σ C lmj · {Σ W} mi · w p (n-
j-i)} ”, the entire right side sums up the arrival signals to the l-th microphone for all speakers.
It represents the total sum of control sounds (secondary sounds) that reach the l-th evaluation point.

Next, the evaluation function (variable to be minimized) J
e

[0048]

[Equation 2]

Let us say.

Then, in order to obtain the filter coefficient W _mi that minimizes the evaluation function Je, the LMS algorithm is adopted in this embodiment. That is, the filter coefficient W _{mi is} a value obtained by partially differentiating the evaluation function Je with respect to each filter coefficient W _mi.
To update.

Therefore, from equation (2),

[0052]

[Equation 3]

From equation (1),

[0054]

[Equation 4]

Therefore, the right side of equation (4) is r
_{If lm} (n−i) is written, the rewriting formula of the filter coefficient is obtained by the following formula (5).

[0056]

[Equation 5]

Here, α is a convergence coefficient and is involved in the speed at which the filter converges optimally and the stability at that time.

Therefore, by performing such control, the noise in the passenger compartment 1 can be reduced.

The present invention is not limited to the above embodiment. For example, even if the evaluation point for noise reduction and the microphone are spatially separated, the residual noise at the evaluation point can be estimated and controlled based on a predetermined value. The control device of the present invention can also be applied to vibration control and the like.

[0060]

As is apparent from the above, according to the configuration of the present invention, it is possible to reduce the noise from a plurality of noise sources with a simple configuration. Moreover, since the reference signal is a combination of the steady definite signal and the random signal, which have no correlation with each other, it is possible to control each independently, and it is possible to reduce noises of various kinds at once.

[Brief description of drawings]

FIG. 1 is a block diagram according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a relationship between a steady decision signal and a random signal.

FIG. 3 is an explanatory diagram showing a noise transmission state.

FIG. 4 is a block diagram according to a conventional example.

[Explanation of symbols]

 3 loudspeaker (control sound source) 5 microphone (residual noise detection means) 7 controller (control means) 50 reference signal creation means

Claims (1)

[Claims] 1. A control sound source for reducing a noise at an evaluation point by generating a control sound that interferes with noise, a means for detecting residual noise at a predetermined position after the interference, and a plurality of noises having no correlation with each other. Means for generating a reference signal by detecting a signal relating to the noise generation state of the source, and control means for outputting a signal for driving the control sound source based on the output signal of the residual noise detecting means and the output signal of the reference signal generating means. An active noise control device comprising: an active noise control device, wherein the reference signal generating means adds a random signal of one kind to at least one steady confirmation signal as a reference signal. ..