JPH0972375A - Active damping device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the generation of vibration by updating the filter factor of an FIR filter so that an addition signal level from an adding part is minimized. SOLUTION: A controller 1 to control operation of an actuator 3 comprises an FIR filter 14 to generate a control waveform signal S (t) having the same frequency and amplitude and opposite phase as those of vibration 5 through regulation of the phase and amplitude of a signal R (t); and a simulation part 7 for updating relating factors. The simulation part 7 comprises a first model forming part 8 to be inputted a signal R' (t) from a sign wave producing part 17; and an adaptive FIR filter 12 to update the filter factor of an FIR filter 14 to which a signal X (t) is inputted. An output signal N (t) from a first adding part 15 and an output signal Y (t) from the adaptive FIR filter 12 are added to generate an error signal e (t). A second addition part 16 is provided to output a signal to the adaptive FIR filter 12 so that the error signal e (t) is minimized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an active vibration damping device for reducing the vibration of a controlled object by adding the vibration having the same frequency and the opposite phase and the same amplitude to the vibration of the controlled object. With regard to the above, the present invention is suitable for use in, for example, damping of steering for steering a vehicle and damping of a vehicle seat.

[0002]

2. Description of the Related Art As an example of this type of prior art, FIG.
An active vibration damping device shown in is known. This technique is based on an actuator 3 that applies an exciting force to the vibration 5 to reduce the vibration 5, a vibration detection sensor 4 that detects the result of the reduction of the vibration 5, and an actuator based on the detection result of the vibration detection sensor 4. The controller 1 for controlling the operation of the control unit 3 and the main unit 3 are configured as follows.

The controller 1 receives the voltage signal D (t) from the vibration detection sensor 4 and the reference signal R (t) from the reference signal source 2 and has the same frequency as the vibration 5.
The control waveform signal S (t) having the same amplitude and opposite phase is output to the actuator 3. This controller 1 uses a reference signal R
FIR that adjusts the phase and amplitude of (t) to form a control waveform signal S (t) having the same frequency, same amplitude, and opposite phase as the vibration 5.
It has a (Finite Impulse Response) filter 14 and a simulation unit 7 for updating the coefficient of the FIR filter 14.

The simulation unit 7 uses a reference signal R
A model forming unit 8 (represented by a symbol of a hat C in the figure) including a FIR filter to which (t) is input, and a model forming unit 8
The adaptive FIR filter 12 which receives the signal X (t) corrected by the above and updates the filter coefficient of the FIR filter 14 is provided. The model forming unit 8 estimates a model of a transfer function H of vibration between the actuator 3 and the vibration detection sensor 4, and corrects the reference signal R (t) to a signal X (t) by the model of this transfer function H. And adaptive F
Input to the IR filter 12. Thus, the reference signal R
The reason why (t) is corrected by the model forming unit 8 is to correct the transmission characteristic of vibration between the arrangement position of the actuator 3 and the arrangement position of the vibration detection sensor 4.

Further, the adaptive FIR filter 12
Inputs the voltage signal D (t) of the vibration detection sensor 4 as an error signal, updates the filter coefficient so as to minimize this error signal, and copies this filter coefficient to the FIR filter 14 to obtain the FIR filter. 14 filter coefficients are updated. In this way, the vibration is reduced by adding the control vibration having the same frequency and the opposite phase and the same amplitude to the vibration.

[0006]

However, in the above-mentioned conventional technique, since the filter coefficient of the FIR filter 14 is used by copying the filter coefficient of the adaptive FIR filter 12, the filter coefficient used by the FIR filter 14 is , A filter coefficient that is delayed by one sampling period from the adaptive FIR filter 12 is used, and as a result, a time delay occurs and there is a problem in that the ability to follow changes in vibration is lacking.

Therefore, according to the present invention, when the vibration of the controlled object is reduced by adding the vibration having the same frequency and the opposite phase and the same amplitude to the vibration of the controlled object. It is an object of the present invention to provide an active vibration damping device that can improve the followability with respect to and reduce vibration.

[0008]

In order to achieve the above object, the technical means according to claim 1 is adopted. According to this technical means, in the addition unit, the output of the second model formation unit,
The output of the vibration detection sensor and the output of the adaptive FIR filter are added. Then, the filter coefficient of the FIR filter is updated so that the level of the added signal output from the adder unit is minimized. As a result, the error caused by the FIR filter using the filter coefficient that is delayed by one sampling period from the adaptive FIR filter is corrected, the followability to the vibration of the controlled object is improved, and the processing accuracy is improved.

According to the second aspect of the present invention, the sine waveform generating section converts the rectangular wave reference signal having a high correlation with vibration into a sine waveform. As a result, since the harmonic components included in the rectangular wave signal are removed, the FIR filter and the adaptive FIR filter only need to process the sine wave, and the processing amount can be reduced and the processing speed can be increased. It will be possible. Further, by reducing the processing amount, it is possible to prevent the divergence of the control and improve the followability to the change of the vibration.

According to the third aspect of the present invention, the sine waveform generator converts the rectangular wave of the internal combustion engine rotation signal of the vehicle equipped with the internal combustion engine into a sine waveform. this is,
For example, a steering vibration component converts a rectangular wave of a rotation signal of the internal combustion engine into a sine waveform because the explosion order component of the internal combustion engine is dominant. Therefore, the vibration of the steering can be effectively reduced by using the rotation signal of the internal combustion engine as the reference signal and converting it into a sine waveform.

According to the fourth aspect of the present invention, the sine waveform generating section converts a rectangular wave of an internal combustion engine rotation signal of a vehicle equipped with an internal combustion engine into a sine waveform. this is,
For example, the vibration component of the seat when the vehicle is in the idling state is based on the fact that the explosion order component of the internal combustion engine is predominant and the rotation signal of the internal combustion engine is used as a reference signal, and is converted into a sine waveform for use. Vibration can be effectively reduced.

According to the fifth aspect of the present invention, the sine waveform generator converts the rectangular wave of the vehicle wheel speed signal into a sine waveform. This is because, for example, the vibration component of the seat when the vehicle is in a running state is dominated by the vibration component synchronized with the rotation of the wheel. Thereby, the vibration of the seat can be effectively reduced.

According to a sixth aspect of the present invention, the controllers are arranged in parallel and a plurality of frequencies are simultaneously active-controlled. As a result, each controller generates control waveform signals corresponding to different frequencies, so that different frequencies can be simultaneously active-controlled. Further, according to the technical means of claim 7, the first and second model forming sections have FIR filters, and
The IR filter length is set so that the number of taps includes only impulses from the observation of the filter coefficient. As a result, the filter length becomes appropriate, and the processing amount can be reduced and the processing speed can be increased.

The actuator may vibrate the yoke and the magnet, and the reaction force of the inertial force generated when vibrating may be used as the exciting force.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the active vibration damping device of the present invention will be described below with reference to the drawings. [First Embodiment] In this embodiment, an active vibration damping device is applied to vibration damping of a steering for steering a vehicle.

FIG. 1 shows a steering wheel 101 of a vehicle 100.
2 is a system configuration diagram of an active vibration damping device applied to the vibration damping of FIG. As shown in FIG. 1, the steering wheel 1
Actuator 3 for giving damping force to the vibration of 01
And a vibration detection sensor 4 for detecting the result of reducing the vibration of the steering wheel 101, and a controller 1 for controlling the operation of the actuator 3 based on the measurement result of the vibration detection sensor 4.

The controller 1 receives a rectangular wave pulse signal R (t), which is an engine rotation signal (not shown), from the crank angle sensor 2 as a reference signal, and converts it into a sine waveform to obtain a signal R '. As (t), a sine waveform generation unit 17 that outputs to the FIR filter (Finite Impulse Response) 14 and the first model formation unit 8 (represented by a hat C in the figure), and a reference signal R ′ (t converted to a sine waveform. ) Is adjusted to form a control waveform signal S (t) having the same frequency, same amplitude, and opposite phase as the vibration 5, and a simulation unit 7 for updating the coefficient of the FIR filter 14. , And an estimation unit 6 for estimating the transfer function of the vibration 5 between the actuator 3 and the vibration detection sensor 4. The reason why the reference signal R (t) is created using the engine rotation signal is based on the finding that the primary component of the engine explosion is the dominant component of the steering vibration.

The estimation unit 6 corrects the control waveform signal S (t) of the FIR filter 14 by the model in which the transfer function of the vibration 5 between the actuator 3 and the vibration detection sensor 4 is estimated. '(Represented by hat C in the figure)
Then, the output of the second model forming unit 8 ′ is inverted and the signal D (t) from the vibration detection sensor 4 is added to the estimated value N of vibration.
And a first addition unit 15 that outputs (t).

On the other hand, the simulation section 7 includes a first model forming section 8 (represented by a hat C in the figure) to which the signal R '(t) from the sine waveform generating section 17 is input, and the first model forming section. The adaptive FIR filter 12 that receives the signal X (t) corrected by the unit 8 and updates the filter coefficient of the FIR filter 14, the output signal N (t) of the first adding unit 15, and the output of the adaptive FIR filter 12 The signal Y (t) is added to form the error signal e (t), and the second adder 16 for outputting the signal to the adaptive FIR filter 12 so as to minimize the error signal e (t). Have. The first adder 15 and the second adder 16 correspond to the adder of the present invention.

FIG. 2 is a configuration diagram showing a mounting state of the active vibration damping device on the steering wheel 101. As shown in FIG. 2, the actuator 3 is connected to the vehicle 10
Shaft 102 of steering 101 for steering 0
The vibration detection sensor 4 is attached to the
It is attached to the actuator 3. The controller 1 is provided in the instrumental panel 103, receives the voltage signal D (t) from the vibration detection sensor 4 and also receives the actuator 3 as described above.
Is configured to output a control waveform signal S (t).

FIG. 3 is a sectional view of the actuator 3. As shown in FIG. 3, the actuator 3 is configured as follows with the permanent magnet 104, the yoke portion 105, and the coil portion 106 as main parts. That is, the yoke portion 105 is held in the case 111 at the neutral position via the spring 107, and
A bearing 109 is provided so as to be movable along the shaft 108. Further, a permanent magnet 104 is arranged on the yoke portion 105, and the permanent magnet 104 is
A coil portion 106 is provided at a position opposed to the coil portion 106 at a predetermined interval.

Then, the permanent magnet 104 and the yoke portion 1 are
In the magnetic circuit composed of 05 and 05, there is a magnetic flux directed toward the center toward the axis 108 as shown by the arrow in the figure, and when the coil portion 106 is energized, the electromagnetic force generated by the Fleming's law moves vertically. Due to this electromagnetic force, the yoke portion 105 and the permanent magnet 104 are used as movable masses to generate vibrations 110, and the reaction force of the inertial force generated at the time of this vibration is used as the exciting force.

Next, FIG. 4 shows the sine waveform generator 1 of FIG.
FIG. 7 is a diagram illustrating a process in which a sine waveform is generated from a rectangular pulse signal R (t) in FIG. The sine waveform generator 17 detects the rising edge of the rectangular pulse signal as shown in FIG. 4A as shown in FIG. 4B, and the intervals t1, t2, t3, ..., Tn of the rising signals. , To be detected. From the interval tn calculated in this way, the sine waveform R '(t) is formed as in Expression 1.

[0024]

## EQU1 ## R '(t) = Asin (2.pi./tn) In this way, since the harmonic components included in the rectangular wave pulse signal R (t) are removed, the FIR filter 14 and the adaptive FIR filter 12 are Sine wave R '(t)
Therefore, it is possible to reduce the amount of processing and to increase the processing speed. Further, by reducing the processing amount, it is possible to prevent the divergence of the control and improve the followability to the change of the vibration.

Next, the control flow of the active vibration damping device having the above-mentioned structure will be described below. FIG. 11 is a flowchart for explaining the overall control algorithm of the controller 1. First, in step 31, it is determined whether or not the model estimation mode is set. What is this model estimation mode?
For example, when an object is placed near the vibration detection sensor 4 or the like, it is executed when the transmission characteristic of vibration between the actuator 3 and the vibration detection sensor 4 changes, and once the transmission characteristic is set. It is judged that the model estimation mode is not necessary while the transfer characteristic does not change. Then, when it is determined that the transfer characteristic has changed and the model estimation mode is necessary, the process proceeds to step 32 and the model estimation mode process is executed.

The process of this model estimation mode will be described with reference to FIGS. 5 and 6. This model estimation mode estimates a model of vibration transfer characteristics between the actuator 3 and the vibration detection sensor 4 in the first and second model forming units 8 and 8 ′, which is shown in FIG. This will be described with reference to the block diagram. As shown in FIG. 5, in addition to the configuration described above, the controller 1 includes a random noise generator 22 that generates a random noise signal S ′ (t) and outputs the random noise signal S ′ (t) to the actuator 3, and random noise from the random noise generator 22. The model estimating unit 25 estimates the model by inputting S ′ (t) and the signal D ′ (t) from the vibration detection sensor 4. The model estimation unit 25 includes a Fourier transform unit, an inverse Fourier transform unit, and other calculation units. Based on this estimation, the first and second model forming units 8,
Although the filter coefficient of the FIR filter forming 8'is determined as follows, the length of the required filter coefficient can be made appropriate at the same time, and accordingly, the amount of calculation can be made appropriate. Note that this estimation is performed after positioning the actuator 3 and the sensor microphone 4.

FIG. 6 shows the model estimating section 25 of FIG.
It is a flowchart which estimates the coefficient of the FIR filter of the 1st and 2nd model formation part 8 and 8 '. Data processing is performed by an A / D converter (Analog to Digital Co
nverter) performed digital data. First, in step S11, the random noise generator 2
At 2, a random noise signal S ′ (t) at time t of the sampling clock is generated.

In step S12, the vibration detection sensor 4 converts the vibration level at that position into a voltage signal D '(t) and inputs it to the controller 1. Then, step S
At 13, the sampling clock time t is compared with the number of data N, and it is determined whether t <N holds. If the determination in step S13 is "YES", the process proceeds to step S14, the sampling clock is increased by one clock, and the process returns to step S11 to steps S11 to S11.
Repeat 13.

On the other hand, in step S13, "NO"
If it is determined, the process proceeds to step S15. In steps S15 to S17 described below, the filter coefficients of the first and second model forming units 8 and 8 ′ are estimated from the transfer function H between the actuator 3 and the vibration detection sensor 4. First, in step S15, the random noise signal S '
Fourier transform of (t) and voltage signal D ′ (t)
Obtain (f) and D (f). Here, the data of (f) is a complex number on the frequency axis.

In step S16, the transfer function H
(F) is calculated based on Equation 2.

[0031]

## EQU00002 ## H (f) = D (f) / S (f) In step S17, the filter coefficients of the first and second model forming units 8 and 8'are estimated from the transfer function H. This estimation is obtained by performing an inverse Fourier transform on H (f) and using the value of the real part as the filter coefficient of the first and second model forming units 8 and 8 ′.

Next, in step S18, the filter coefficients of the first and second model forming sections 8 and 8'are observed, as will be described later, and only the impulse, which is a characteristic part of the filter coefficient, is included. Determine the number of taps. The random noise generator 22 and the model estimation unit 2 shown in FIG.
5 does not need to be integrated with the system of the controller 1 of FIG. 1, and may be provided separately.

Next, FIG. 7 is a diagram for explaining the determination of the number of FIR filters of the first and second model forming units 8 and 8'obtained by the model estimating unit 25 (details of step S18). The first and second model forming parts 8 shown in FIG.
The 8 ′ FIR filter coefficient has the data number of 1024, 4
The transfer function H is obtained by the Fourier transform of the 00 line, and the inverse Fourier of the transfer function H is obtained. In the shape of such a filter, the number of taps can be reduced to 25 because the impulse is sufficiently contained up to the number of taps of 25. As described above, the first and second model forming units 8 and 8 '
The number of taps of the FIR filter is determined in advance, and FIR filters having a small number of taps are formed as the first and second model forming units 8 according to the flowchart of FIG. As a result, an appropriate filter length can be determined in advance, and the processing amount can be reduced and the processing speed can be increased.

FIG. 8 is a flow chart showing another embodiment for estimating the FIR filter coefficient of the first and second model forming sections 8 and 8 '. In this case, as the model estimating unit 25 in FIG. 5, the model for estimating the transfer function H between the actuator 3 and the vibration detection sensor 4 is LMS (Le
ast Mean Square) adaptive FIR filter is used.

First, in step S61, the random noise generator 22 generates a random noise signal S '(t) at time t of the sampling clock. In step S62, the vibration detection sensor 4 converts the vibration level at that position into a voltage signal D ′ (t), and the controller 1
To enter. Then, in step S63, the adaptive FIR filter of the model estimation unit 25 sets the time t
The output Y ′ (t) of the FIR filter is calculated by using Expression 3 using the filter coefficient C ′ (i) of

[0036]

(Equation 3) In step S64, the error e ′ (t) is set to D ′.
The difference between (t) and Y ′ (t) is calculated as in Expression 4.

[0037]

## EQU00004 ## e '(t) = D' (t) -Y '(t) In step S65, the filter coefficient is updated, and the filter coefficient at time t + 1 is calculated as in Expression 5.

[0038]

## EQU5 ## ^{C't + 1} (i) = ^C't (i) -2 × α _c ×
S ′ (t−i) × e ′ (t) where α _c is a step size parameter. In step S66, the condition for ending the estimation is determined. This estimation condition satisfies that the change amount of the filter coefficient is less than or equal to a certain value, the absolute value average of the error e ′ (t) is less than or equal to a certain value, and the processing time is more than or equal to a time sufficient for the convergence of the filter coefficient. is there. If this condition is not met, step S6
If the condition is satisfied, the process ends. Step S67
In step S, advance the sampling clock by one
Proceed to 61 and repeat the same procedure thereafter.

The FIR filter having the filter coefficient thus obtained forms the first and second model forming units 8 and 8 which form a model for estimating the transfer function between the actuator 3 and the vibration detection sensor 4. 'Becomes. Note that the number of FIR filters showing the model of the transfer function is determined based on the impulse portion by observing as described in detail in FIG. 7.

Returning to FIG. 11, after the model estimation mode processing is completed, this main routine is terminated. On the other hand, if it is determined in step 31 that the mode is not the model estimation mode, the process proceeds to step 33 and the active control mode process is executed. The process of the active control mode will be described with reference to FIGS. FIG. 9 is a flowchart explaining the active control mode of the active noise controller 1 of FIG.

First, the model estimating method of the estimating unit 6 will be described, and then the active control will be described. FIG.
And 10 are active noise controllers 1 of FIG.
3 is a flowchart (Nos. 1 and 2) for explaining the active control mode of FIG. First, in step S41, the reference signal R (t) at time t and the output D of the vibration detection sensor 4
Input (t) to the controller 1. Then, the sine waveform generation unit 17 generates a sine waveform R ′ (t) based on the reference signal R (t).

In step S42, the vibration estimating section 6 estimates the vibration 5. That is, the control waveform signal S
The signal d (t) obtained by passing (t) through the second model forming unit 8 ′.
Is calculated as in Equation 6.

[0043]

(Equation 6) In step S43, the signal D (t) 10 from the vibration detection sensor 4 and the vibration d from the second model forming unit 8 '.
The estimated value N (t) of vibration is calculated based on Equation 7 using the difference from (t).

[0044]

## EQU00007 ## N (t) = D (t) -d (t) In step S44, the reference signal R (t) is input to the simulation unit 7 for updating the coefficient. In the simulation unit 7, the signal X (t) is created through the first model forming unit 8 based on the equation 8 as follows.

[0045]

(Equation 8) Then, the signal X (t) obtained by the mathematical formula 8 is input to the adaptive FIR filter 12. This adaptive F
In the IR filter 12, as described above, in the LMS algorithm, in order to update the P filter coefficients W (i) using the error series e (t), the FI is calculated in step S45.
The output Y (t) of the R filter is created as in Expression 9.

[0046]

[Equation 9] In step S46, the sum of the previously calculated estimated value N (t) of vibration and Y (t) is calculated as the error e (t) as in Expression 10.

[0047]

[Mathematical formula-see original document] e (t) = N (t) + Y (t) = D (t) + {Y (t) -d (t)} Then, in step S47, the filter coefficient is calculated as in Expression 11. Update.

[0048]

[Equation 11] W ^{t + 1} (i) = W ^t (i) -2 × α _W × X
(Ti) * e (t) Here, i: 0 to P-1. In step S48, the filter coefficient of the adaptive FIR filter 12 is copied to the FIR filter 14 as shown in Expression 12.

[0049]

## EQU12 ## W ^{t + 1} (i) → W (i) where i: 0 to P-1. In step S49, the reference signal R (t) is passed through the FIR filter 14 to create the control waveform signal S (t) as shown in Expression 13.

[0050]

(Equation 13) Then, in step S50, the control waveform signal S
(T) 9 is output to the actuator 3.

Next, in step S51, it is determined whether to continue the process. If the determination is "YES", the process proceeds to step S52, and if "NO", the process is terminated. In step S52, the sampling clock is advanced by one, and the process proceeds to step S41 to synchronize with the sampling clock.
These processes are repeated. As described above, according to the present embodiment, in step S46, {Y (t)-
d (t)} represents the error of the FIR filter 14 when the filter coefficient is delayed by one sampling period, and this is added and corrected as the error signal of the adaptive FIR filter 12, so that the vibration of the steering 101 is The following capability is improved and the processing accuracy can be improved. [Second Embodiment] In the present embodiment, an active vibration damping device is applied to vibration damping of a vehicle seat.

FIG. 13 is a system configuration diagram of an active vibration damping device applied to damping the seat 150 of the vehicle 100. As shown in FIG. 13, the actuator 3 that applies a damping force to the vibration of the seat 150, the vibration detection sensor 4 that detects the result of reducing the vibration of the seat 150, and the measurement result of the vibration detection sensor 4 The controller 1 controls the operation of the actuator 3 based on the above. The actuator 3 and the vibration detection sensor 4 are embedded in the seat 150.

A rectangular wave pulse signal R (t), which is a rotation signal of the engine 155, is input from the crank angle sensor 2 to the controller 1, and a rectangular wave which is a rotation signal of the wheel is received from the wheel speed sensor 160. Pulse signal r (t) is input. And the controller 1
When the vehicle 100 is in the idling state, the sine waveform generation unit 17 of sine waveform R ′ based on the pulse signal R (t) of the rectangular wave from the crank angle sensor 2
(T) is generated, and when the vehicle 100 is in a traveling state, a rectangular wave pulse signal r from the wheel speed sensor 160 is generated.
A sine waveform R '(t) is generated based on (t). The reason for this is that the vibration component of the seat 150 when the vehicle 100 is in the idling state is dominated by the explosion order component of the engine, and the seat 150 when the vehicle 100 is in the running state.
The vibration component transmitted to is based on the finding that the component synchronized with the rotation speed of the wheels of the vehicle is dominant.

Since the configuration and control contents of the controller 1 are the same as those described in the first embodiment, the description thereof is omitted here. As described above, according to the present embodiment, the error caused by the FIR filter 14 using the filter coefficient that is delayed by one sampling period from the adaptive FIR filter 12 is corrected, and the followability to the vibration of the sheet 150 is good. Therefore, the processing accuracy is improved. [Third Embodiment] FIG. 14 is a view showing the arrangement of an active vibration damping system according to the third embodiment. In the present embodiment, when the vehicle 100 equipped with the 4-cylinder engine is in the idling state, the seat 150 generated due to the rotational secondary component (first explosion) and the rotational fourth component (secondary explosion) of the engine 155. It is a system that simultaneously controls the vibration of.

In the present embodiment, the controller 1 is provided with a control unit 180 corresponding to the rotational secondary component (first explosion) and a control unit 190 corresponding to the rotational fourth component (secondary explosion). A rectangular wave pulse signal R (t), which is an engine rotation signal, is input from the crank angle sensor 2 to the control units 180 and 190, and a sine waveform generation unit 17-1 of the control unit 180 outputs a rotational secondary component. A sine waveform of (explosion primary) is generated, and a sine waveform generation unit 17-2 of the control unit 190 generates a sine waveform of a rotational fourth-order component (explosion secondary). The configurations and control contents of the control units 180 and 190 are the same as those of the first and second control units described above.
This is the same as the embodiment.

Further, the controller 1 is provided with a third adder 200, and the signals output from the FIR filters 14-1, 14-2 of the controllers 180, 190 are added by the third adder 200. The control waveform signal S (t) is generated, and this control waveform signal S (t) is input to the actuator 4. According to this embodiment, the FIR filters 14-1, 1
4-2 is an adaptive FIR filter 12-1, 12-
The error caused by using the filter coefficient that is delayed by one sampling period from 2 is corrected, the followability to the vibration of the sheet 150 is improved, the processing accuracy is improved, and the plurality of rotation order components (rotation of the engine 155 (rotation) are It is possible to simultaneously reduce the vibration of the sheet 150 due to the secondary component and the rotational quaternary component).

By combining the third embodiment with the second embodiment, the rectangular wave pulse signal R (t) of the crank angle sensor 2 when the vehicle 100 is in the idling state. Based on the above, the vibration of the seat 150 caused by the rotational second-order component and the rotational fourth-order component of the engine 155 is reduced, and when the vehicle 100 is in a traveling state, a rectangular wave pulse signal r ( The vibration of the seat 150 may be reduced based on t).

[Brief description of drawings]

FIG. 1 is a diagram showing a system configuration in which an active vibration damping device according to a first embodiment of the present invention is applied to damping a steering wheel.

FIG. 2 is a configuration diagram showing a mounting state of the active vibration damping device of FIG. 1 on a steering wheel.

FIG. 3 is a cross-sectional configuration diagram showing an actuator that constitutes the active vibration damping device of FIGS. 1 and 2.

FIG. 4 is a diagram for explaining calculation of a sine waveform period from an input rectangular wave in the sine waveform generation unit of FIG.

5 is a block diagram illustrating model estimation of a transfer characteristic H between the actuator 3 and the vibration detection sensor 4 in the first model forming unit 8 of FIG.

FIG. 6 is a model forming unit 8 by the model estimating unit 25 of FIG.
3 is a flowchart for estimating the FIR filter coefficient of FIG.

FIG. 7 is a diagram illustrating the determination of the number of FIR filters of the first model forming unit 8 obtained by the model estimating unit 25.

FIG. 8 is a flowchart illustrating another example of estimating the FIR filter coefficient of the model forming unit 8.

FIG. 9 is a flowchart (part 1) explaining an active control mode in the controller 1 of FIG.

10 is a flowchart (No. 2) explaining the active control mode in the controller 1 of FIG.

FIG. 11 is a flowchart illustrating an overall control algorithm in the controller 1 of FIG.

FIG. 12 is a diagram showing a conventional active vibration damping device.

FIG. 13 is a diagram showing a system configuration in which an active vibration damping device according to a second embodiment of the present invention is applied to vibration damping of a seat.

FIG. 14 is a diagram showing a system configuration in which an active vibration damping device according to a third embodiment of the present invention is applied to vibration damping of a seat, and vibrations of the seat corresponding to a plurality of rotational order components of the engine are damped. is there.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 controller 3 actuator 4 vibration detection sensor 8, 8'first and second model forming unit 12 adaptive FIR filter 14 FIR filter 15 first adding unit 16 second adding unit 17 sine waveform generating unit 25 model estimating unit

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Kuninaga Matsuoka 14 Iwatani, Shimohakaku-cho, Nishio-shi, Aichi Japan Auto Parts Research Institute

Claims (8)

[Claims] 1. An actuator that outputs an exciting force for reducing vibration, a vibration detection sensor that measures the vibration reduction effect, and active control of the actuator based on a detection signal of the vibration detection sensor. In the active vibration damping device including a controller, the controller is configured to output a control waveform signal that adjusts a phase and an amplitude of a reference signal that is highly correlated with the vibration to the actuator.
An IR filter, an adaptive FIR filter for updating the filter coefficient of the FIR filter, a transfer characteristic of vibration between the actuator and the vibration detection sensor is estimated, and the reference signal is corrected by this transfer characteristic, A first model forming unit that outputs a corrected signal to the adaptive FIR filter, a transfer characteristic of vibration between the actuator and the vibration detection sensor is estimated, and the output of the FIR filter is corrected based on this transfer characteristic. 2 model forming section, and an adding section for adding the output of the second model forming section, the output of the vibration detection sensor, and the output of the adaptive FIR filter, wherein the adaptive FIR filter outputs from the adding section. The FIR so that the added signal level is minimized.
An active vibration damping device characterized by updating a filter coefficient of a filter. 2. The active vibration damping device according to claim 1, wherein the controller further includes a sine waveform generation unit that converts a rectangular wave signal having a high correlation with the vibration into a sine waveform. Active vibration control device. 3. The active vibration damping device according to claim 2, which is used for steering a vehicle equipped with an internal combustion engine, wherein the sine waveform generation unit converts a rectangular wave of an internal combustion engine rotation signal of the vehicle into a sine waveform. Active damping device characterized by the above. 4. The active vibration damping device according to claim 2, wherein the active vibration damping device is used for a seat of a vehicle equipped with an internal combustion engine, wherein the sine waveform generation unit converts a rectangular wave of an internal combustion engine rotation signal of the vehicle into a sine waveform. Active damping device characterized by the above. 5. The active vibration damping device according to claim 2, wherein the sine waveform generation unit changes the rectangular wave of the wheel speed signal of the vehicle into a sine waveform. 6. An active vibration damping device, wherein the controllers according to claim 2 are arranged in parallel, and a plurality of frequencies are simultaneously active-controlled. 7. The active vibration damping device according to claim 1, wherein the first and second model forming units have FIR filters, and the filter length of the FIR filters is a filter coefficient of a filter coefficient. From the observation, the active vibration damping device is characterized in that the number of taps includes only impulses. 8. The active vibration damping device according to claim 1, wherein the actuator vibrates a yoke and a magnet with an electromagnetic force to obtain the exciting force. Shaking device.