JPH05241583A - Active type noise controller - Google Patents

Abstract

PURPOSE:To provide a device muffling actively a noise consisting of plural spectra generated by an engine by outputting a second sound. CONSTITUTION:The rotational speed of the engine 1 is detected by a sensor 7 and converted into a voltage by an F/V converter 11 and inputted to a processor part 16 from an A/D 12. Sine wave data stored in the ROM of the processor part 16 previously changes whose reading interval (plural pieces exist) in accordance with the voltage value and is outputted as a reference signal. The reference signal is processed by an adaptive filter and is outputted as the second sound from speakers 5, 6 through a D/A 14, a power amplifier 15. The adaptive filter is updated successively so that the sound pressure of a microphone becomes minimum. Then, since the reading speed of the sine wave data is made variable and plural reference signals are made, plural spectra noises are muffled simultaneously with a simple system.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention has the same amplitude for noise,
The present invention relates to an active noise control device that cancels noise by generating a secondary sound of opposite phase, and particularly relates to generation of a reference signal.

[0002]

2. Description of the Related Art Active noise control that cancels noise with waves of the same amplitude and opposite phase has been used in the 1950s since the pioneering research (USP-2043416, 1936) by Lueg in the 1930s.
It can be said that this is an old technology that has been studied by n, Conver, etc., but it has only recently been actually applied to products. This is largely due to the advent of high-speed arithmetic elements for enabling control such as digital signal processors, but it can also be mentioned that the theoretical aspects of control algorithms have been improved.

As recent recent noteworthy examples of active noise control technology, there are two examples, one by GBB Chaplin (for example, Japanese Patent Publication No. 56-501062) and one by PANelson / SJ Elliot (for example, Japanese Patent Publication No. 1-501344). I can name it. The difference between the two control methods is that the former control uses "repetitive control" that assumes the periodicity of the target noise, whereas the latter control uses the LMS algorithm, which is a type of steepest descent method. The target noise does not necessarily need to be periodic because the adaptive signal processing is performed.

This LMS adaptive control algorithm is a method codified by Widrow in the 1960s (eg, B. Widro).
w / PAMantey / BBGoode "Adaptive Antenna Systems,
(PROCEEDINGS OF THE IEEE, Vol.55, NO.12, DEC 1967)
However, because of its versatility, recent research examples on active noise control mostly rely on this control algorithm. Also in the present invention, it is basically premised on the use of this control algorithm.
The prior art will be described by taking 1344 (PANelson / SJ Elliot) as an example.

FIG. 12 shows an active noise control device 12100 described in the above-mentioned published patent, in which a plurality of loudspeakers and microphones muffle a particular closed space such as a vehicle interior. It has three microphones 1212 that measure the sound pressure at a given position in a closed space.
And primary sound (noise) at each microphone 1212 position and 2
Two loudspeakers 11 for outputting secondary sounds for reducing noise by interference of secondary sounds, a reference signal (reference signal) generator 1215 for generating a signal 124 synchronized with the rotation of the engine 122, a reference It is composed of a control circuit 1213 having a pair of adaptive filters 1214 for driving the loudspeaker 1211 by phase-amplitude modulating the signal 124 to output a signal 123 for driving the loudspeaker. Further, an engine rotation signal (for example, an ignition timing signal, a crank angle sensor signal, etc.) is input to the reference signal generator 1215, and the reference signal generator 1215 is proportional to an integral multiple of the engine rotation cycle at every moment. Generating a sine wave signal.

The LMS adaptive control algorithm constantly updates the coefficient of the adaptive filter 1214 so that the square value of the sound pressure at each microphone 1212 position is minimized, but the primary sound and the secondary sound interfere well. Therefore, in order to reduce noise, the standard signal 124 or the reference signal 1216 which is the source of the standard signal 124 must include a component having a sufficiently high correlation with the primary sound. Usually, a dimensionless quantity that takes a value between 0 and 1 called coherence is defined as an index showing the degree of correlation between two signals. From the result of rigorous theoretical analysis, it is known that the noise reduction amount by the active noise control system based on the LMS adaptive control algorithm is determined by the value of this coherence.

In the active noise control system in the passenger compartment of an automobile as shown in FIG. 12, the noise caused by the rotational vibration of the engine is the object of control, and the engine rotation signal is supplied as a reference signal and synchronized with this. By generating the sine wave signal, the reference signal having a high coherence with respect to the engine noise component is obtained. For example, the engine is 4
In the case of a four-cycle cylinder, vibration with a frequency twice that of engine rotation is large, and is generally called engine secondary vibration. This is due to gas torque fluctuation due to gas combustion every 1/2 rotation and inertia torque fluctuation due to imbalance of moment of crankshaft system.
Then, this becomes an exciting torque to vibrate the vehicle body, and this is propagated into the vehicle interior to generate standing wave noise. Therefore, at this time, the active noise control of the secondary vibration noise becomes possible by supplying the rotation signal for each crank angle of 180 degrees as the reference signal.

[0008]

By the way, the LMS adaptive control algorithm can be applied not only when noise of a single spectrum is silenced as shown in FIG. 12 but also when a plurality of noise spectra are included. is there.

For this purpose, it is necessary that the reference signal contains spectral components having a sufficiently high coherence at the same frequency for each noise spectral component.

The aforementioned published patents present several methods for generating the reference sinusoidal signal.

First, a rotation pulse train synchronized with the engine frequency is passed through a tracking filter having the engine frequency as a fundamental frequency to obtain the Nth engine speed.
It is to convert to a signal that includes harmonics at a constant rate.

Secondly, a rotation pulse train (128 pulses per rotation, for example) sufficiently larger than the engine speed is divided to generate a plurality of rotation pulse trains having different frequencies, and only the fundamental frequency of each pulse signal is passed. This is a method of passing the signals through a tracking filter and synthesizing the filtered signals.

The third method is to measure the time interval of the rotating pulse train using a clock pulse generator and generate a sine wave having a period equal to the pulse interval by a digital oscillator. In this way, by including the sinusoidal wave signals of a plurality of frequency components in the reference signal or the reference signal, it is possible to muffle the noise spectra of a plurality of corresponding rotation order components at the same time.

By the way, in industrial vehicles, for example, medium to small trucks of several tons or less, 4 to 4 are used as power sources.
A diesel engine of about 6 cylinders is usually used, but in addition to the engine, auxiliary equipment such as a generator and a hydraulic pump and a transmission are built in, and these all rotate in synchronization with the engine. In trucks, the engine is usually mounted slightly behind the cabin, and the vibration of the engine itself travels through the body to excite the entire cabin and generate standing wave noise, as well as the rotation of auxiliary machinery and transmission. Since all vibrations also generate noise by exciting the cabin, the cabin room is filled with noise composed of a plurality of spectral components proportional to the engine rotation speed. Further, the magnitude of the noise spectrum is not always large only in a certain component, for example, the secondary vibration component, and changes depending on various situations.

In such a machine, the reference signal must include not only the engine rotational order component but also a spectral component of a frequency that is proportional to the engine rotational speed but is not necessarily an integral multiple of its fundamental order. Yes, and again LM
From the viewpoint of the adaptability of the S adaptive control algorithm, it is desirable to make the ratio of the magnitudes of the respective spectral components variable depending on the situation.

However, as described above, the first and second
In the method of generating the pulse train having the reference frequency of the engine speed and passing the pulse train through the tracking filter to obtain the reference signal as in the above example, there are the following problems.

In order to mute the spectrum component having a frequency that is a non-integer multiple of the fundamental order of engine rotation, for example, 7
In order to mute the spectrum component of / 4 times the frequency, prepare a circuit that generates 28 pulses per rotation, divide this frequency, and generate a pulse that contains the spectrum component of the 7/4 times frequency. To do. As described above, it is difficult to generate a pulse including a non-integer multiple frequency because it is necessary to generate a large number of pulses.

No consideration is given to making the ratio of the magnitudes of the respective spectral components included in the reference signal variable. In the example above, the reference signals at multiple frequencies are
After passing through different tracking filters, they are combined and become the input of the speaker. Therefore, in order to make the ratio of the magnitudes of the respective spectral components included in the reference signal variable, the characteristics of the tracking filter must be changed, which is difficult.

Further, in the method using the digital oscillator from the time interval of the rotating pulse train like the third method,
In order to generate the trigonometric function, a series-expanded expression representing the trigonometric function is calculated at high speed inside the processor to obtain the trigonometric function. For this reason, there is a problem that the load on the processor is increased to simultaneously generate sine waveforms having a plurality of frequencies.

It is an object of the present invention to provide an active noise control device capable of easily muting a spectrum component having a frequency which is a non-integer multiple of the fundamental order.

[0021]

In order to solve the above problems, in an active noise control device for canceling noise by generating a secondary sound of the same amplitude and opposite phase to the noise, the noise for detecting the noise Detecting means, secondary sound outputting means for outputting a secondary sound that interferes with noise at the evaluation point, and detecting means for detecting a velocity signal depending on the frequency of the periodic motion (vibration or rotation) of the noise source. And, the speed signal is input,
A reference signal generation unit (also referred to as a reference signal detection unit) that outputs a reference signal that serves as a reference for secondary sound generation, and an input signal to the secondary sound output unit based on the detected noise and the reference signal. The reference signal generating means has a signal processing means for determining, and the reference signal generating means changes a reading speed of the waveform data according to the value of the speed signal and a storage means for storing the reference waveform data. And a waveform output means for reading the waveform data and outputting it as a reference signal.

[0022]

In the active noise control device for canceling the noise by generating the secondary sound having the same amplitude and opposite phase to the noise, the noise detecting means detects the noise. The secondary sound output means outputs a secondary sound that interferes with the noise at the evaluation point. The detection means detects a velocity signal that depends on the frequency of the periodic motion of the noise source. The signal processing means determines an input signal to the secondary sound output means based on the detected noise and the reference signal. The storage means included in the reference signal generation means stores reference waveform data. The waveform output means changes the read speed of the waveform data according to the value of the speed signal, reads the waveform data, and outputs the read waveform data as a reference signal.

[0023]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The contents of the present invention will be described in detail below with reference to the drawings. FIG. 1 shows an embodiment in which an active noise control device is applied in order to reduce noise in a cabin space of a truck, which is a typical industrial vehicle.

In the truck 50 shown in the figure, a four-cylinder diesel engine 1 as a power source is attached to the rear of the cabin 2, and auxiliary equipment such as a generator and a hydraulic pump that are driven by the engine 1 to rotate. And a transmission (not shown) are attached. And, all of the vibrations caused by them vibrate the cabin 2 and become noise,
The interior of the cabin is filled with noise composed of a plurality of spectral components proportional to the engine speed.

In the figure, the noise inside the cabin is operated.
A system is shown in which sound pressure is detected by two microphones 3 and 4 arranged near the ear of a speaker, and a secondary sound is generated by two speakers 5 and 6 to mute the sound. A tacho sensor 7 for generating a pulse signal of 4 pulses per rotation is attached to a rotary shaft that rotates once per 2 rotations of the crankshaft of the diesel engine 1, and a signal 1 of 2 pulses per 1 rotation of the crank.
01 is obtained. Control for active noise control
The unit 10 is an F unit for inputting the signal of the tacho sensor 7.
S / V converter 11, A / D converter 12, A / D converter 13 for inputting sound pressures 102, 103 measured by microphones 3, 4, secondary sound control signals 104, 105 , D / A converter 1 for driving 6
4, a power amplifier 15, and a processor unit 16 that executes LMS adaptive control using these input signals.

Here, the F / V converter 11 inputs the rotation pulse signal 101 every moment and outputs an analog voltage 107 proportional to the rotation speed of the crankshaft. And
The analog voltage 107 is converted into a digital voltage 106 via the A / D converter 12 and input to the processor section 16. FIG. 2 shows an example of the circuit configuration of the F / V converter 11 using a general-purpose linear IC (RC4151). RC4151 is an IC that operates with a single power supply. Input frequency f _IN
The output voltage V ₀ proportional to can be obtained by the equation shown in the figure.

Next, FIG. 3 shows an example of the configuration of the processor section 16. Here, the processor unit 16 is a single-chip microcomputer 20 of about 8 bits for creating a reference signal.
And a digital signal processor (DSP) 30 that executes a high-speed product-sum operation for adaptive control. The microcomputer 20, the CPU 21, the RAM 22,
It comprises a ROM 23, I / O ports 24 and 25, a counter 26, etc., while a DSP 30 has a CPU 31, R
AM32, ROM33, I / O ports 34, 35, 36
Etc. are built in.

FIG. 4 is a diagram showing the operation of the microcomputer 20 of FIG. A process of generating a digital reference signal will be described with reference to FIGS. 3, 4 and 5. ROM is shown in FIG.
It is the data X _i for one cycle of the sine wave written in 23. This is because the amplitude of the sine wave (Y axis) is 8 bits (-1
27 to 127), the angle (X axis) is 8 bits (0 to 25)
5), and when the waveform one cycle is 360 degrees, the X axis is every 360 / 256≈1.4 °. The 256 8-bit data (256 bytes) are sequentially read by the CPU 21 at a certain fixed timing.

FIG. 4B shows a digital voltage 106 obtained from the A / D converter 12 via the I / O port 25 for the reading speed (time interval) of the sine waveform data in the ROM 23.
The flow chart more sought is shown. First, the value V ₀ of the digital voltage 106 is calculated by the CPU from the I / O port 25.
Then, the read time interval t _R is calculated by the following equation.

[0030]

## EQU1 ## t _R = B / V ₀ However, B is a constant stored in the ROM 23, and this value is preset so that t _R × 256 is one period of the noise spectrum to be silenced. (Since the division requires time, if this time is not allowed, the value of t _R corresponding to the digital voltage value V may be stored in the ROM and read.) FIG. 5 shows a flowchart for controlling the reading of ROM data by using the built-in counter 26. That is, the counter 26 operates independently, clocks the clock signal generated by the built-in clock generator, and updates its value every moment (FIG. 5 (a)). If the read interval t _R is set at a certain time, the value t of the counter 26 at that time is set.
_C = T _C is loaded into the CPU 21, and ΔT _C = T _C + t _R is calculated and temporarily stored in the RAM 22. The value t _{C of the} counter 26 is constantly monitored and compared with ΔT _C. Then, when t _C ≧ ΔT _C , an interrupt signal is generated again. Then, again, the value of the counter 26 at that time is the CPU 2
It is loaded to 1, a new ΔT _C is calculated, and the same operation is repeated.

On the other hand, when the interrupt signal is received, the CPU 2
1 has a sinusoidal digital data X ₀ in the ROM 23.
Is loaded to the I / O port 24 and the I / O port 24
It is sent to the / O port 34 (FIG. 5 (b)). When the interrupt signal is generated again, the next data X ₂ is DSP.
Sent to 30. In this way, digital data X ₀ , X ₁ , X ₂ , ..., X ₂₅₅ , X ₀ , ... Of sine waveforms are sent to the DSP 30 _every moment and used for control as reference signals. Then, if the read interval t _R is changed, it is possible to deal with a noise component having an arbitrary frequency.

Now, the data X _i sent to the DSP 30 at a certain time is loaded into the CPU 31 and is stored in the RAM.
It is stored in 32 as the reference signal X (n). here,
FIG. 6 is a flow chart of the operation showing the control sequence of the LMS adaptive control algorithm written in the ROM 33.
It is

First, the CPU 31 causes the filter coefficients W ₁₁ , W ₁₂ , W ₂₁ , W ₂₂ of the adaptive filter and the reference signal X.
(N), X b from (n-1) to RAM 32 - De and output Y ₁ (n), Y ₂ (n) is calculated by the following equation (step 1).

[0034]

[Equation 2]

Here, the number of taps of the adaptive filter is at most 2.
The tap is nothing but a single frequency for both the reference signal and the output signal. The value of the adaptive filter is initialized to a proper value, for example 0.

The outputs Y ₁ and Y ₂ of the calculation result are output from the speakers 5 and 6 through the D / A converter 14 and the power amplifier 15 as the output signals 104 and 105 shown in FIG. It becomes a secondary control sound (step 2).

On the other hand, the sound pressures e ₁ , e ₂ (102, 103) measured by the microphones 3, 4 are input to the CPU 31 from the I / O port 35 via the A / D converter 13. (Step 3) At this time, the evaluation function J = e ₁ ² + e ₂ ²
Then, each filter coefficient of the adaptive filter is updated at constant sampling intervals so as to minimize J. This is given by the minimum value when J is differentiated by each filter coefficient.

Here, the sound pressures e ₁ and e ₂ are the primary sound (noise) d.
_It is given by the following equation as the sum of ₁ , d ₂ and the secondary sound.

[0039]

[Equation 3]

However, C _lmj is the acoustic transfer coefficient between the speaker and the microphone, and is represented by the tap number j.
Substituting equation 2 into equation 3 yields the following equation.

[0041]

[Equation 4]

Generally, in updating the filter coefficient in the LMS adaptive control algorithm, (∂J / ∂W _m ) is used and given by the following equation.

[0043]

[Equation 5]

Here, μ is generally called a convergence coefficient and is related to the speed of adaptation.

When J is calculated from equation 4 and (∂J / ∂W _m ) is calculated, the updating formula of each filter coefficient is shown by the following formula.

[0046]

[Equation 6]

However, α = 2μ. The CPU 31 accesses the data necessary for the calculation from the RAM 32 at any time, and
Is executed at high speed (step 4).

The operations of (step 1) to (step 4) are executed at constant sampling intervals, and at the time of the next sampling, the operation returns to (step 1) again and the same control sequence is executed. (Here, the sampling time for creating the reference signal by the microcomputer 20 and the sampling time for updating the adaptive filter by the DSP 30 do not necessarily need to be synchronized.) In the embodiment shown in FIGS. For example, it is an active noise control system for only the rotational secondary vibration component. However, in this system, by preparing a plurality of reading speeds for the waveform data, a plurality of noise components having different frequency components can be silenced at the same time. For example, a truck, a construction machine vehicle, or the like has a feature that a noise level generated by an element such as a generator, a transmission, or a hydraulic pump that rotates in synchronization with an engine is large. When the rotational speed of the engine changes, the rotational speeds of these driven elements also change, but the rotational speed ratio is constant and the noise frequency ratio is also constant. Therefore, in the formula for calculating the readout interval t _R of the waveform data shown in Equation 1, if the value of the constant A is set according to the frequency ratio, an arbitrary frequency doubled signal can be obtained with respect to the engine rotation frequency. You can

FIG. 7 shows, as an active noise control system for simultaneously muting a plurality of noise spectra, rotational 8th order, 7 / 2nd order rotations produced by the rotational second-order and fourth-order components of a four-cylinder diesel engine and driven elements. It shows a system that silences the components simultaneously. The digital voltage value V _* obtained through the A / D converter 12 is loaded into the CPU 21 of the microcomputer 20 (step 11), and the time interval for reading the sine waveform data in the ROM is calculated by the following equation. ..

[0050]

[Equation 7]

However, A ₁ , A ₂ , A ₃ , and A ₄ are constants for producing sinusoidal waveforms having frequencies of 4, 4, 8 and 7/2 times the engine rotation frequency ( Step 1
2).

When the read intervals t _R1 , t _R2 , t _R3 and t _R4 obtained by the equation 7 are set at a certain time point, the value T _C of the counter 26 at that time point is obtained. (Step 1
3) The following equation is calculated from the read interval and the value of the counter to determine the data read timing. (Step 1
4)

[0053]

[Equation 8]

The counter value t _C is updated every moment, and the value is constantly compared with ΔT _R1 , ΔT _R2 , ΔT _R3 , ΔT _R4 (in the above case, ΔT _R3 <ΔT _R2 <ΔT _R4 <ΔT _R1 ). ing. Then, when t _C first becomes equal to or greater than ΔT _R3 , an interrupt signal is generated and data X _i (i = 0 ... 2) is generated.
55) are sequentially loaded and the signal train X3 = {..., X3 _i , ...} Is stored in the RAM 22. On the other hand, the counter value T _C3 at that time is read and ΔT _R3 = t _R3 + T _C3 is reset (step 15).

Similarly, the counter value t _C is ΔT _R2 ,
The values of ΔT _R4 and ΔT _R1 are also compared in order, and when the values are equal to or larger than the respective values, an interrupt signal is generated and the data is deleted.
Data X _i are sequentially loaded. Then, the signal sequence X2 = {...,
X2 _i , ...}, X4 = {..., X4 _i , ...}, X1 = {..., X1 _i ,
...} is created and stored in the RAM 22. Then, the counter values T _C2 , T _C4 , and T _C1 at that time are read and ΔT
_R2 = t _R2 + T _C2 , ΔT _R4 = t _R4 + T _C4 , ΔT _R1 = t _R1 +
T _C1 is reset (steps 16-18). on the other hand,
Since the digital voltage value V _* is updated every moment according to the change in the rotation speed of the crankshaft, V _* is reloaded at a constant timing and a series of control sequences are performed again.
(Return to step 11) The four signal trains thus obtained are sent from the I / O port 24 of the microcomputer 20 and used as reference signals in the DSP 30. The output time intervals of are different. Generally, the sampling interval of the digital filter may be twice the input frequency (Nyquist frequency). Therefore, the signal sequences Χ1, X2, X3, and X4 are sampled at a constant time that is at least twice the maximum frequency X3 (8 times the engine speed), and the reference signal sequence S ₁ = {..., S1 at constant time intervals. _i ,
_{...,}, S 2 = {} ..., S2 i, ...}, S 3 = {..., S3 i, ...},
Consider finding S ₄ = {..., S 4 _i , ...}. FIG. 8 is a flowchart showing this constant sampling method. First, the constant sampling interval t _S is obtained from V _* , then the counter value T _CS at the time when t _S is set is read, and the read timing ΔT _S = T _CS + t _S is calculated.
The counter values t _C and ΔT _{S, which} are updated every moment, are constantly compared, and when t _C ≧ ΔT _S , an interrupt signal is generated and R
The data X1 _i , X stored in the AM22 at that time
2 _i , X3 _i , and X4 _i are loaded into the CPU 21, and the RAM 22 is again used as the reference signal sequences S1 _i , S2 _i , S3 _i , and S4 _i at fixed time intervals.
Stored in. Then, the value T of the counter at that time
_CS is read and repeated in the same manner.

Next, the data stored in the RAM 22 is loaded again into the CPU 21 and the DS from the I / O port 24.
It is sent to the I / O port 34 of P30. Then, it is temporarily stored in the RAM 32 via the CPU 31. These are loaded into the CPU as respective reference signal sequences S _k (k = 1 to 4) at a predetermined sampling time, and each has a 2-tap adaptive filter coefficient W _kim (k = 1 to 4, m = 0, 1). Executes the sum of products operation with S _k (n), and outputs y _km (n)
Is obtained as follows.

[0057]

[Equation 9]

However, m = 1, 2 and indicates the speaker number. Each output y _km (n) obtained by the above equation is added for each subscript m for the subscript k.

[0059]

[Equation 10]

The outputs Y ₁ and Y _{2 are} low-pass filters,
It is output as a secondary sound from the speaker via a power amplifier or the like. The LMS adaptive control algorithm for generating the secondary sound so as to minimize the sound pressures e ₁ and e ₂ measured by the microphone is basically the same as the control sequence shown in FIG. 6 in the first embodiment. Next, in the sampling method shown in FIG. 8, the adaptive filter is operated independently for each of a plurality of reference signal sequences. On the other hand, FIG. 9 shows a method in which a plurality of generated signal sequences (herein referred to as reference signals) are added to generate one reference signal and the adaptive filter is operated only for this reference signal. .. That is, FIG.
Reference signal S1 obtained by sampling for a fixed time as in
The arithmetic mean of _i , S2 _i , S3 _i , and S4 _i is calculated from time to time using the following equation.

[0061]

[Equation 11]

Then, {..., SA _i , ...} obtained momentarily is R
The data is stored in the AM 22, and then the data is loaded again to the CPU 21 and sent to the DSP 30. The DSP 30 uses the adaptive filter W as a reference signal sequence.
_A1i, W _A2i (the number of taps of this adaptive filter only must be set to the length can all frequency components included in the reference signal representation) is filtered by, spin - output signal to mosquito follows Is asked.

[0063]

[Equation 12]

In Equation 11, the sampling signal S
The reference signal was obtained by simple addition averaging of 1 _i , S2 _i , S3 _i , and S4 _i . When the reference signal contains sinusoidal waves of multiple frequencies, such as the noise in the cabin of a truck, the volume of the corresponding noise spectrum is also affected by the amplitude ratio of each reference signal component. .. Therefore, FIG.
Shows a case in which a reference signal is created by performing weighted averaging among these respective reference signals, and the ratio of the weights is made variable according to the operating conditions of the engine. In this case, Equation 11 is rewritten as the following equation.

[0065]

[Equation 13]

However, a ₁ , a ₂ , a ₃ , a ₄ are weight variables of each reference signal.

The truck 50 is provided with means for detecting changes in the engine speed and changes in the acoustic transmission characteristics of the control space (in the cabin) (opening and closing of windows, changes in temperature and humidity). Further, in the ROM 22 of the microcomputer 20, the values of a ₁ , a ₂ , a ₃ , a ₄ corresponding to these are written as a map. Now, a certain characteristic change-for example, when the acoustic transfer characteristic greatly changes, such as when the cabin door is opened, the composition ratio of each component of the noise spectrum also changes.
In such a case, for example, the open / closed state of the door is detected by some sensor (for example, an infrared sensor), and a signal is sent to the microcomputer 20. The controller receiving the signal selects the value of the weight variable corresponding to the state from the ROM table,
The values of the variables a ₁ , a ₂ , a ₃ , and a ₄ are changed, and the addition operation for creating the reference signal of Expression 13 is executed.

Unlike the above embodiment, the active noise control device has a plurality of filters, and the waveform output means reads the waveform data according to the value of the speed signal. At this time, reading is performed at a plurality of reading rates having different magnifications, a plurality of waveform data are simultaneously output, each is input to a different filter, and a plurality of reference signals having different periods are output and output. It is also possible to add items that have been added.

In the active noise control device, the speed signal is not converted into a voltage signal, but a pulse signal is used as the speed signal, and the reference signal detecting means directly detects the time interval of the pulse signal. Then, the reading speed of the waveform data may be varied according to the time interval.

Further, as an active noise controller unit for canceling noise, reference signal detecting means for receiving a speed signal and outputting a reference signal, and the detected secondary noise based on the detected noise and the reference signal. The reference signal detection means has a storage means for storing reference waveform data and a read speed of the waveform data that is variable according to the value of the speed signal. And a waveform output means for reading the waveform data and outputting it as a reference signal, and a microphone, a speaker and a sensor can be arbitrarily combined.

By the way, the noise in the cabin of the truck as shown in FIG. 1 is a vibration radiated sound generated by the vibration of the engine transmitted through the body to vibrate a certain portion in the cabin such as a panel. Therefore, as a noise reduction method, the generated noise is not canceled by using the secondary sound of a speaker or the like, but the vibration radiated sound itself is generated by giving the panel vibration of the same amplitude and opposite phase as the vibration transmitted from the engine. It is also possible to suppress it. FIG. 11 is a diagram showing a silencing control method for vibrating radiated sound by such active vibration. Detection of engine rotation signal by tacho sensor 7, microphone 3,
4 is the same as that shown in FIG. 1 in the detection of the sound pressure signal in the cabin and the configuration of the control unit 10, but it is replaced by the loudspeakers 5 and 6 and the actuator for vibration is placed at an appropriate position in the cabin 2. -55, 56 are attached. The actuators 55 and 56 are controlled by the control unit 10 by the LMS adaptive control algorithm so that the sound pressure detected by the microphones 5 and 6 is minimized.

In the above, a truck, which is a typical example of an industrial vehicle driven by a four-cylinder diesel engine, is taken as an example, and a system for actively muting noise in the cabin has been described with reference to the drawings. This active noise control device is not limited to this because it is a general-purpose technology due to its nature, and it has construction machines such as passenger cars and hydraulic excavators, and other rotating machines such as engines and motors. It can be applied anytime when the space where noise is generated is silenced.

As described above, according to the present invention, the microcomputer R
Since the sine wave data stored in advance in the OM is used as the reference signal by changing the reading speed according to the engine rotation speed, high-speed and highly-accurate active noise control can be realized with a simple system. ..

By preparing a plurality of ROM data read speeds and simultaneously generating a plurality of reference signals,
Noises consisting of multiple spectra can be silenced at the same time. Further, by adding these plural reference signals into one reference signal, active noise control of plural spectrum noises can be realized by a simple system. In addition, even if the acoustic transfer characteristics greatly change and the composition ratio of each component of the noise spectrum changes, it is possible to quickly mute the sound by using a system in which the weighting for each signal during signal addition is made variable. Is.

In the present invention, the sine wave data to be stored is not limited to one cycle, but may be half cycle, two cycles, or the like.

Further, regarding the speaker, it is not necessary to provide a dedicated speaker for noise control, and for example, a speaker for audio may also be used.

[0077]

According to the present invention, it is possible to provide an active noise control device that can easily mute a spectral component having a frequency that is a non-integer multiple of the fundamental order.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of an active noise control device.

FIG. 2 is a circuit configuration diagram of an F / V converter.

FIG. 3 is a block diagram of a processor unit

[Figure 4] Operation diagram of the microcomputer

FIG. 5 is an explanatory diagram of a method of reading ROM data.

FIG. 6 is a control sequence diagram of adaptive control.

FIG. 7 is an explanatory diagram of a ROM reading method when muting multiple spectra.

FIG. 8 is an explanatory diagram of a constant sampling method when muting multiple spectra.

FIG. 9 is an explanatory diagram of a reference signal creation method by signal addition.

FIG. 10 is an explanatory diagram of a reference signal creation method by weighted addition.

FIG. 11 is an explanatory diagram of silencing control of vibration radiation sound by active excitation.

FIG. 12 is a block diagram of a conventional active noise control device.

[Explanation of symbols]

1 ... Engine, 2 ... Cabin, 3, 4 ... Microphone, 5, 6 ... Speaker 7 ... Tacho sensor, 10 ... Control unit, 11 ...
F / V converter, 12, 13 ... A / D converter, 14 ...
D / A converter, 15 ... Power amplifier, 16 ... Processor section, 101 ... Crank rotation signal, 102, 103 ... Microphone sound pressure 104, 105 ... Secondary sound control signal, 106 ... Digital voltage

Claims (9)

[Claims] 1. An active noise control device for canceling noise by generating a secondary sound of the same amplitude and opposite phase to the noise, the noise detecting means for detecting the noise and the noise at the evaluation point interfering with each other. A secondary sound output means for outputting a secondary sound to be caused, a detection means for detecting a signal depending on the frequency of the periodic motion of the noise source, and a reference for generating the secondary sound by inputting the speed signal. The reference signal generating means for outputting a reference signal; and the signal processing means for determining an input signal to the secondary sound output means on the basis of the detected noise and the reference signal. Storage means for storing reference waveform data, and changing the reading speed of the waveform data according to the value of the speed signal to read the waveform data and output it as a reference signal. And a waveform output means. That active noise control system. 2. The active noise control device according to claim 1, wherein the reference signal generating means receives a number of pulses proportional to the rotation speed of the rotating machine and generates a voltage proportional to the rotation speed. A proportional voltage generating means and a means for converting the voltage into a digital value by an A / D converter, the digital value being a speed signal, and the storage means having a normalized digital amplitude, at constant intervals. The active noise control device is characterized in that the waveform data of the sine wave sampled in (1) is stored, and the waveform output means reads and outputs the waveform data. 3. The active noise control device according to claim 1 or 2, wherein the waveform output means, when reading the waveform data according to the value of the speed signal, provides a plurality of different magnifications. The active noise control device is characterized in that reading is performed at a reading speed that it has, a plurality of waveform data are simultaneously output, and a plurality of reference signals having different periods are output. 4. The active noise control device according to claim 1, wherein the waveform output means, when reading the waveform data in accordance with the value of the speed signal, provides a plurality of different magnifications. The active noise control is characterized in that it has signal adding means for performing reading at a read speed, outputting a plurality of waveform data at the same time, adding the plurality of waveform data, and outputting the added waveform as a reference signal. apparatus. 5. The active noise control device according to claim 4, wherein when the waveform output means outputs a plurality of sine waveforms and the signal addition means generates the added waveform as a reference signal,
An active noise control device characterized by weighting and adding the amplitude of each waveform. 6. An active noise control device for exciting a specific position with respect to noise to cancel the noise, comprising noise detecting means for detecting the noise, and secondary exciting means for exciting the specific position. Detecting means for detecting a velocity signal depending on the frequency of periodic motion of a noise source; reference signal generating means for receiving the velocity signal and outputting a reference signal serving as a reference for generating a secondary sound; A signal processing unit that determines an input signal to the secondary vibration unit based on the generated noise and the reference signal, and the reference signal generation unit stores a reference waveform data. Means for changing the reading speed of the waveform data according to the value of the speed signal, reading the waveform data, and outputting the waveform data as a reference signal. Type noise control device. 7. The active noise control device according to claim 1, wherein the speed signal is a pulse signal, and the reference signal generating means is a time interval of the pulse signal. Is detected and the reading speed of the waveform data is changed according to the time interval. 8. The active noise controller according to claim 1, 2, 3, 4 or 5, wherein the secondary sound output means is a speaker and the noise detection means is a microphone. Active noise control device. 9. A detected noise and a velocity signal depending on the frequency of vibration or rotation of a noise source are input and interfered with the noise at the evaluation point with the same amplitude and opposite phase with respect to the noise. An active noise controller unit for determining a secondary sound and canceling noise, comprising: a reference signal generating means for receiving the speed signal and outputting a reference signal serving as a reference for generating a secondary sound; The reference signal generating means has a signal processing means for determining the secondary sound based on noise and the reference signal, the reference signal generating means stores a reference waveform data, and a value of the speed signal. An active noise controller unit comprising: a waveform output means for changing the reading speed of the waveform data in response to the reading of the waveform data and outputting the waveform data as a reference signal.