JP2012168282A - Active-type oscillation noise suppression apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an active-type oscillation noise suppression apparatus in which oscillation noise suppression effects can be surely presented in the vicinity of a resonance frequency of a first transfer system even if this frequency is included within a frequency range of a controlled object.SOLUTION: In the case where a transfer function of a first transfer system is substituted into a substitution transfer function Q that is a state where a negative feedback element P is added to a first true transfer function G using a pole arrangement method, on the basis of a first estimated transfer function Gh, the negative feedback element P is calculated for substituting a true resonance frequency fin the first transfer system into a frequency fother than a controlled object frequency range. On the basis of a first filter coefficient calculated by a substituted estimated transfer function Qh, a first control signal y1 is generated. On the basis of the first control signal y1 and the negative feedback element P, a second control signal y2 is generated. Then, a third control signal y3 is generated by subtracting the second control signal y2 from the first control signal y1, and the third control signal y3 is outputted as a control signal.

Description

  The present invention relates to an active vibration noise suppression apparatus that can actively suppress vibration and noise using adaptive control.

  As an apparatus that actively suppresses vibration and noise at an evaluation point using adaptive control, there is one described in Patent Document 1.

JP-A-8-44377

  In order to suppress vibration and noise at the evaluation point, it is important to accurately estimate the amplitude and phase of the actual transfer function G of the first transfer system (hereinafter referred to as “first actual transfer function G”). is there. In particular, if there is a difference between the phase of the actual first actual transfer function G and the phase of the estimated value Gh of the transfer function of the first transfer system (hereinafter referred to as “first estimated transfer function Gh”), the evaluation is performed. There is a possibility that vibration and noise at the point cannot be suppressed.

  By the way, a resonance frequency exists in any transmission system. Near the resonance frequency of the transmission system, the phase of the transmission system changes greatly. Therefore, when the vicinity of the resonance frequency of the first transmission system is included in the frequency range to be controlled, it is necessary to estimate the phase of the first actual transfer function G in the vicinity of the resonance frequency of the first transmission system with particularly high accuracy. .

  However, even if the phase of the first actual transfer function G in the vicinity of the resonance frequency of the first transmission system can be estimated with high accuracy in the initial state, the resonance frequency of the first transmission system may change due to changes over time, temperature changes, etc. Change. For this reason, after the change, there is a possibility that a difference occurs between the actual phase of the first actual transfer function G and the phase of the first estimated transfer function Gh. As a result, vibration and noise at the evaluation point may not be appropriately suppressed.

  For this reason, it is common to limit the frequency range of the controlled object so that the vicinity of the resonance frequency of the first transmission system is not included in the frequency range of the controlled object. However, there is a demand for expanding the frequency range to be controlled. Then, the vicinity of the resonance frequency of the first transmission system may be included in the frequency range to be controlled.

  The present invention has been made in view of such circumstances, and even when the resonance frequency of the first transmission system is included in the frequency range to be controlled, the vibration and noise suppression effect is reliably ensured in the vicinity of the frequency. An object of the present invention is to provide an active vibration and noise suppression device capable of exhibiting the above.

An active vibration noise suppression device of the present invention is an active vibration noise suppression device that generates control vibration or control sound according to a control signal and actively suppresses vibration or noise at an evaluation point. In a first transmission system up to the evaluation point through a frequency calculation unit that calculates a frequency of a noise generation source and a control vibration control sound generator that generates the control vibration or control sound after outputting the control signal Using the first transfer function estimated value calculation unit for calculating the estimated value Gh of the first actual transfer function G and the pole placement method, the transfer function of the first transfer system is negatively fed back to the first actual transfer function G. When the replacement transfer function Q is added with the feedback element P added, the actual resonance frequency f _G0 in the first transfer system is set to the control target frequency range based on the estimated value Gh of the first actual transfer function G. Non-frequency Based on the feedback element calculation unit for calculating the negative feedback feedback element P to be replaced with the number f _Q0 , the estimated value Gh of the first actual transfer function G and the negative feedback feedback element P, the replacement transfer function Q Based on the replacement transfer function estimated value calculation unit for calculating the estimated value Qh of the first transfer coefficient, the first filter coefficient calculated by the estimated value Qh of the replacement transfer function Q, and the frequency calculated by the frequency calculation unit. A first control signal generator for generating a control signal y1, a second control signal generator for generating a second control signal y2 based on the first control signal y1, the negative feedback feedback element P, and the frequency; A third control signal y3 is generated by subtracting the second control signal y2 from the first control signal y1, and the third control signal y3 is output as the control signal. A third control signal generation unit.

According to the present invention, the actual resonance frequency f _G0 of the first transmission system is replaced with another frequency f _Q0 using the pole placement method. That is, the resonance frequency of the first transmission system apparently becomes the frequency _fQ0 . That is, when viewed virtually, the resonance frequency _fQ0 of the first transmission system can be excluded from the frequency range to be controlled. As a result, since it is considered that the resonance frequency is not included in the frequency range to be controlled virtually, even if the actual phase of the first actual transfer function G and the phase of the first estimated transfer function Gh greatly deviate, It is possible to suppress a significant shift between the phase of the replacement transfer function Q and the estimated value Qh of the replacement transfer function. Therefore, vibration and noise at the evaluation point can be reliably suppressed. In particular, even if the first transmission system changes due to changes over time or temperature, vibration and noise can be reliably suppressed.

  The replacement transfer function estimated value calculation unit outputs the third control signal y3 obtained by subtracting the second control signal y2 from the first control signal y1 as the control signal. In addition, the estimated value Qh of the replacement transfer function Q may be calculated based on vibration or sound at the evaluation point based on the third control signal y3.

  Since the estimated value Qh of the replacement transfer function Q is measured by actually generating control vibration or control sound, the estimated value Qh of the replacement transfer function Q can be calculated with higher accuracy. It should be noted that the estimated value Qh of the replacement transfer function Q is obtained based on the estimated value Gh of the first actual transfer function G and the negative feedback feedback element P only by calculation without actually generating control vibration or control sound. be able to.

  In addition, the first transfer function estimated value calculation unit, when the third control signal generation unit outputs the first control signal y1 as the control signal, the vibration at the evaluation point by the first control signal y1 or The estimated value Gh of the first actual transfer function G is updated based on the sound, and the feedback element calculation unit determines the negative feedback feedback element P based on the updated estimated value Gh of the first actual transfer function G. You may make it update.

Since the estimated value Gh of the first actual transfer function G is sequentially updated, even if the first actual transfer function G changes due to secular change or the like, the estimated value Gh of the first actual transfer function G is the current first It can follow changes in the actual transfer function G. Since the negative feedback feedback element P is updated based on the updated estimated value Gh of the first actual transfer function G, the negative feedback feedback element P becomes a value corresponding to the current first actual transfer function G. . Therefore, the resonance frequency _fQ0 of the first transmission system can be reliably positioned outside the control target frequency range.

  The active vibration noise suppression device further includes a residual signal detection unit that detects a residual signal due to interference between vibration or noise caused by the generation source and the control vibration or control sound at the evaluation point, and the first control The signal generation unit includes a first filter coefficient calculated from the estimated value Qh of the replacement transfer function Q, the frequency calculated by the frequency calculation unit, and the residual signal detected by the residual signal detection unit. Based on this, the first control signal y1 may be generated by an adaptive control method.

Virtually, adaptive control can be performed in a state where the resonance frequency _fQ0 of the first transmission system is not included in the frequency range to be controlled. As a result, even if the phase of the first actual transfer function and the estimated value Gh of the first actual transfer function are shifted due to changes over time, vibration or noise can be reliably suppressed.

  In addition, the first control signal generation unit may perform the feedforward control method based on the first filter coefficient calculated by the estimated value Qh of the replacement transfer function Q and the frequency calculated by the frequency calculation unit. The first control signal y1 may be generated.

Virtually, map control (feed-forward control) can be performed in a state where the resonance frequency _fQ0 of the first transmission system is not included in the frequency range to be controlled. Thereby, vibration or noise can be reliably suppressed.

It is a control block diagram before applying the pole placement method. It is a control block diagram after applying the pole placement method. It is an actual control block diagram when the pole placement method is applied. A first example of frequency characteristics depending on whether or not the pole placement method is applied will be shown. The 2nd example of the frequency characteristic by the presence or absence of application of the pole placement method is shown. It is a control block diagram of the active vibration noise suppression apparatus when applied to adaptive control. FIG. 7 is a deemed view of FIG. 6. It is an actual functional block diagram of the active vibration noise suppression device when applied to adaptive control. It is a flowchart which shows the calculation update process of Gh map data, P data, and Qh map data. It is a functional block diagram regarding the calculation update process of Gh map data and the negative feedback feedback element P. It is a functional block diagram regarding the calculation update process of Qh map data. It is a control block diagram of the active vibration noise suppression device when applied to map control. FIG. 13 is a deemed diagram of FIG. 12. It is an actual functional block diagram of the active vibration noise suppression apparatus when applied to map control.

<1. Control concept applying the pole placement method>
A control block diagram before applying the pole placement method will be described with reference to FIG. As shown in FIG. 1, consider a case where the vibration noise generation source 10 generates vibration or noise and is transmitted to the evaluation point 20 via the second actual transfer function H. Then, the first control signal generation unit 120 outputs the first control signal y1 and generates the control vibration or control sound corresponding to the first control signal y1, thereby suppressing the vibration or noise at the evaluation point 20. For the purpose.

  That is, as shown in FIG. 1, at the evaluation point 20, the vibration or noise generated by the vibration noise generation source 10 is transmitted via the second actual transfer function H, and the first control signal y1. The control vibration or control sound Z transmitted through the first actual transfer function G interferes. Temporarily, at the evaluation point 20, the vibration or noise X generated by the vibration noise generation source 10 is transmitted via the second actual transfer function H, and the first control signal y1 is the first actual transfer function G. When the control vibration or control sound Z transmitted through the control line completely coincides, there is no vibration or noise at the evaluation point 20.

  Then, as shown in FIG. 2, the first actual transfer function G of the first transfer system is replaced with a replacement transfer function Q. That is, state feedback control is performed on the first actual transfer function G. Specifically, the negative feedback feedback element P is fed back from the output side of the first actual transfer function G to the input side and subtracted from the first control signal y1. In this way, the pole placement method is used to replace the transfer function of the first transfer system with the replacement transfer function Q shown in FIG. 2 from the first actual transfer function G shown in FIG. By using the pole placement method, the eigenvalue of the transfer function of the first transfer system can be changed. Changing the eigenvalue changes the resonance frequency.

With reference to FIGS. 4A and 4B and FIGS. 5A and 5B, amplitude and phase frequency characteristics in the case where the pole placement method is not applied and in the case where the pole placement method is applied will be described. . Here, as shown in FIGS. 4A and 4B, the amplitude of the first actual transfer function G is _AG , the phase is Φ _G , the amplitude of the replacement transfer function Q is A _Q , and the phase is Φ _Q And

As shown in FIGS. 4A and 4B, by applying the pole placement method, the actual resonance frequency f _G0 of the first actual transfer function G can be changed to the resonance frequency f _Q0 of the replacement transfer function Q. Even if the actual resonance frequency f _G0 of the first actual transfer function G is included in the control target frequency range (f _{a to} f _b ), the resonance frequency f _Q0 of the replacement transfer function Q is outside the control target frequency range. can do. Then, as shown in FIG. 4 (b), the control frequency range of interest, the change width of the phase [Phi _Q substituent transfer function Q is compared to the variation of the phase [Phi _G of the first actual transfer function G, significantly It can be seen that there is a reduction.

Here, FIG. 4 (a) (b) showed a state in which changed the actual resonance frequency f _G0 to a frequency lower than the lower limit value f _a control target frequency range of the first real transfer function G. In addition, as shown in FIGS. 5A and 5B, the actual resonance frequency f _G0 of the first actual transfer function G can be changed to a frequency higher than the upper limit value f _b of the control target frequency range. Closer of the lower limit value f _a and the upper limit value f _b of the real resonance frequency f _G0 is controlled target frequency range of the first real transfer function G, it is preferable to be changed.

  However, it is difficult to detect the control vibration or the control sound Z transmitted from the first control signal y1 via the first actual transfer function G in FIG. Therefore, as shown in FIG. 3, a negative feedback feedback element P is generated from the input side of the first actual transfer function G. However, in order to obtain an equivalent control state, an estimated value Gh of the first actual transfer function G (hereinafter referred to as “first estimated transfer function Gh”) is arranged on the inlet side of the negative feedback feedback element P. For convenience of description, the estimated value “hat (＾)” is described as “h” in the text of the specification.

  That is, the first actual transfer function G is identified, and the first estimated transfer function Gh is calculated. Then, the feedback element P is calculated by applying the pole placement method using the first estimated transfer function Gh. And as shown in FIG. 3, the active vibration noise suppression apparatus 100 outputs the 3rd control signal y3 as a control signal. The third control signal y3 is a signal obtained by subtracting the second control signal y2 from the first control signal y1. The second control signal y2 is a signal transmitted to the third control signal y3 via the first estimated transfer function Gh and the negative feedback feedback element P. In this way, the third control signal y3 can be generated.

<2. Explanation of the pole placement method>
Next, a method for calculating the negative feedback feedback element P will be described using the pole placement method. Suppose that the eigenvalue poles of the current system are λ1 _real and λ2 _real and that the eigenvalue poles are to be the target values λ1 _tar and λ2 _tar . The eigenvalue poles λ1 _real and λ2 _real− of the current system are shown in equation (1), and the eigenvalue poles λ1 _tar and λ2 _tar of the target system are shown in equation (2).

Here, τ _tar and ε _tar are as shown in Equation (3).

At this time, the characteristic equation with the target eigenvalue poles λ1 _tar and λ2 _tar is expressed as shown in Equation (4).

  By the way, as shown in FIG. 2, the equation of motion when the second control signal y2 is applied is expressed by equation (5). This equation (5) can be replaced as in equation (6).

  Here, assuming that the second control signal y2 depends on the displacement x and the speed v, the second control signal y2 is expressed as in Expression (7). That is, the negative feedback feedback element P is expressed by P1 and P2 as shown in Expression (8).

  Substituting equation (7) into equation (6) yields equation (9).

  The characteristic equation of the control system obtained from this equation is expressed as equation (10), and becomes equation (11).

  In order for Formula (4) and Formula (11) to have the same pole, P1 and P2 become Formula (12). That is, the negative feedback feedback element P can be obtained from Expression (12).

<3. Active vibration and noise suppression device when applied to adaptive control>
Next, the active vibration noise suppression device 100 when the control concept described with reference to FIG. 3 is applied to adaptive control will be described with reference to FIGS.

(3.1) Control Block Diagram First, a control block diagram of the active vibration noise suppression device 100 will be described with reference to FIGS. 6 and 7. The adaptive filter coefficient updating unit 180 constituting the active vibration noise suppression apparatus 100 calculates the update amount ΔW of the adaptive filter coefficient W using the estimated value Qh of the replacement transfer function Q and the residual signal e at the evaluation point 20. The adaptive filter coefficient W is composed of an amplitude filter coefficient a and a phase filter coefficient φ.

  Then, the first control signal generation unit 120 generates the first control signal y1 based on the adaptive filter coefficient W and the frequency f. The second control signal generation unit 130 generates the second control signal y2 when the third control signal y3 is transmitted via the first estimated transfer function Gh and the negative feedback feedback element P. Here, the first estimated transfer function Gh and the negative feedback feedback element P are sequentially calculated.

  The third control signal generation unit 140 generates a third control signal y3 obtained by subtracting the second control signal y2 from the first control signal y1. Then, the third control signal generation unit 140 outputs the third control signal y3 as a control signal. The output third control signal y3 is transmitted to the evaluation point 20 via the first actual transfer function G. At the evaluation point 20, the control vibration or control sound Z to which the third control signal y3 is transmitted via the first actual transfer function G and the vibration or noise generated by the vibration noise generating source 10 are the second actual transfer function H. The vibration or the noise X transmitted through the antenna interferes.

  Here, the region surrounded by the alternate long and short dash line in FIG. 6 corresponds to the replacement transfer function Q. That is, FIG. 6 can be expressed as shown in FIG. According to FIG. 7, at the evaluation point 20, the control vibration or control sound Z to which the first control signal y <b> 1 is transmitted via the replacement transfer function Q and the vibration or noise generated by the vibration noise generation source 10 are second actual. The vibration or noise X transmitted via the transfer function H interferes.

That is, it can be considered that the transfer function of the first control system is replaced with the replacement transfer function Q from the first actual transfer function G. Even if the actual resonance frequency f _G0 of the first actual transfer function G is included in the control target frequency range, the resonance frequency f _Q0 of the replacement transfer function Q is obtained by using the pole placement method described above. It can be changed outside the range. That is, the phase [Phi _Q substituent transfer function Q does not vary significantly in the control frequency range of interest. Therefore, even if the phase Φ _{G of} the first actual transfer function G and the phase Φh _{G of} the first estimated transfer function Gh are greatly shifted, the phase Φ h of the phase Φ _Q of the replacement transfer function Q and the estimated value Qh of the replacement transfer function Q It is possible to suppress a large deviation from _Q. Therefore, vibration and noise at the evaluation point 20 can be reliably suppressed in the control target frequency range. In particular, even if the first transmission system changes due to changes over time or temperature, vibration and noise can be reliably suppressed.

(3.2) Functional Block of Entire Control A case where the above-described control block diagram is applied as an actual functional block will be described with reference to FIG. Here, as an actual functional block, an automobile will be described as an example. In an automobile, it is desired that the engine (internal combustion engine) 10 becomes a vibration noise generation source so that vibration and noise generated by the engine 10 are not transmitted to the vehicle interior. Therefore, in order to actively suppress vibrations and noise (such as suppression target vibrations) generated by the engine 10, control generators and the like are generated by the generator 150. In the following description, the active vibration and noise suppression device 100 is described as an example of a device that is applied to an automobile and suppresses vibration or noise generated by the engine 10, but is not limited thereto. The present invention can be applied to anything that generates vibration and noise to be suppressed.

  This active vibration noise suppression apparatus 100 includes a frequency calculation unit 110, a first control signal generation unit 120, a second control signal generation unit 130, a third control signal generation unit 140, a generation device 150, a residual signal, A detection unit 160, a replacement estimated transfer function data selection unit (hereinafter referred to as "Qh data selection unit") 170, an adaptive filter coefficient update unit 180, a Gh map data storage unit 210, a P data storage unit 220, Qh map data storage unit 230.

  The frequency calculation unit 110 receives a periodic pulse signal from a rotation detector (not shown) that detects the number of revolutions of the engine 10, and based on the pulse signal, generates vibration or noise (suppression target) generated by the engine 10. The frequency f of the main component of vibration etc. is calculated.

The first control signal generation unit 120 generates the first control signal y1 _(n) as a sine wave obtained according to the equation (13) based on the frequency f calculated by the frequency calculation unit 110 by adaptive control. Here, the subscript _(n) is a subscript representing the sampling number (time step). That is, as apparent from the equation (13), the first control signal y1 _(n) includes the frequency f, the amplitude filter coefficient a _(n) as the adaptive filter coefficient W _(n) , and the phase filter coefficient φ _(n). Is a signal at time t _(n) . Then, the amplitude filter coefficient a _(n) and the phase filter coefficient φ _(n) are adaptively updated by an adaptive filter coefficient update unit 180 described later.

The second control signal generation unit 130 transmits the third control signal y3 _(n−1) generated by the third control signal generation unit 140 via the first estimated transfer function Gh and the negative feedback feedback element P. The second control signal y2 _(n) in the case where it is assumed is generated. Here, the first estimated transfer function Gh is selected based on Gh map data stored in a Gh map data storage unit 210 (to be described later) and the frequency f calculated by the frequency calculation unit 110. Further, the negative feedback feedback element P is stored in a P data storage unit 220 described later.

The third control signal generation unit 140 generates a third control signal y3 _(n) obtained by subtracting the second control signal y2 _(n) from the first control signal y1 _(n ) as shown in Expression (14). Then, the third control signal generation unit 140 outputs the third control signal y3 _(n) as a control signal. The third control signal y3 _(n) is used by the second control signal generator 130 as described above.

The generator 150 (corresponding to the “control vibration control sound generator” in the present invention) is an apparatus that actually generates vibration and sound. The generator 150 is driven based on the third control signal y3 _(n) output by the third control signal generator 140. For example, the generator 150 that generates the control vibration is, for example, a vibration generator arranged in a frame or a subframe (not shown) connected to the drive system. The generator 150 that generates the control sound is, for example, a speaker. In the case where the generator 150 is a device that generates a control vibration or control sound using, for example, magnetic force, the current, voltage, or power supplied to a coil (not shown ₎ is used as the third control signal at each time t _(n) . By controlling so as to respond to y3 _(n) , the generator 150 generates a control vibration or control sound corresponding to the third control signal y3 _(n) .

Then, at the evaluation point 20, the control vibration or the like Z _(n) transmitted by the generator 150 via the transmission system B and the suppression target vibration or the like generated by the engine 10 are the second. The vibration noise X _(n) transmitted through the actual transfer function H is synthesized.

Therefore, the residual signal detection unit 160 is arranged at the evaluation point 20 and detects residual vibration or residual noise (corresponding to “residual signal” in the present invention) e _(n) at the evaluation point 20. This residual vibration e _(n) is expressed by equation (15). For example, an acceleration sensor or the like can be applied as the residual signal detection unit 160 that detects the residual vibration e _(n) . Further, as the residual signal detection unit 160 that detects the residual sound e _(n) , a sound absorbing microphone or the like can be applied. The ideal state is that the residual signal e _(n) detected by the residual signal detector 160 becomes zero. The first actual transfer function G is a transfer function of the transfer system from the third control signal generator 140 that outputs the third control signal y3 _(n) to the evaluation point 20 through the generator 150. is there. That is, the first actual transfer function G includes the transfer function of the generator 150 itself and the transfer function of the transfer system B between the generator 150 and the evaluation point 20.

The Qh data selection unit 170 selects a replacement estimation transfer function Qh corresponding to the frequency f calculated by the frequency calculation unit 110 from the map data of the replacement estimation transfer function Qh stored in the Qh map data storage unit 230. To do. The replacement estimated transfer function Qh is stored in the map data of the replacement estimated transfer function Qh. The replacement estimated transfer function Qh is represented by an amplitude component Ah _Q and a phase component Φh _Q corresponding to the frequency f, as shown in Expression (16). In Equation (16), the replacement estimated transfer function Qh, the amplitude component Ah _Q, and the phase component Φh _Q are in accordance with the frequency f. (F), Ah _Q (f) and Φh _Q (f). This replacement estimated transfer function Qh is used in an adaptive filter coefficient updating unit 180 described later.

The adaptive filter coefficient updating unit 180 adaptively updates the adaptive filter coefficient W _(n) for configuring the first control signal y1 _(n) described above. The adaptive filter coefficient W _(n) is composed of an amplitude filter coefficient a _(n) and a phase filter coefficient φ _(n) as shown in Expression (17).

The adaptive filter coefficient updating unit 180 applies adaptive control using an adaptive least mean square filter (Filtered-X LMS). Here, adaptive filter coefficient updating section 180 updates adaptive filter coefficient W so as to minimize evaluation function J set based on residual signal e _(n) . Hereinafter, how to derive an update expression for updating the adaptive filter coefficient W in the adaptive filter coefficient updating unit 180 will be described.

The evaluation function J is defined as in equation (18). That is, the evaluation function J is the square of the residual signal e detected by the residual signal detection unit 160. A control signal y _(n) that minimizes the evaluation function J is obtained.

Next, the gradient vector _(n) is calculated according to the equation (19). The gradient vector ▽ _(n) is obtained by partial differentiation of the evaluation function J _(n) with the adaptive filter coefficient W _(n) . Then, the gradient vector ▽ _(n) is represented as the right side.

The adaptive filter coefficient W _{(n + 1)} is derived by subtracting the term obtained by multiplying the gradient vector _（(n) thus calculated by the step size parameter μ from the previously updated adaptive filter coefficient W _(n) . When the expression is expanded in this way, it is expressed as Expression (20).

Here, the adaptive filter coefficient W _(n) is composed of the amplitude filter coefficient a _(n) and the phase filter coefficient φ _(n) as shown in the equation (17). That is, the update formula of the amplitude filter coefficient a _(n) is expressed as shown in Expression (21), and the update expression of the phase filter coefficient φ _(n) is expressed as shown in Expression (22). Here, (1 / Ah _Q ) in Expression (21) is obtained by adding a normalization process to the update of the amplitude filter coefficient a _(n) .

That is, the adaptive filter coefficient updating unit 180 calculates the previously updated amplitude filter coefficient a _{(n) based} on the replacement estimation transfer function Qh, the residual signal e _(n), and the amplitude step size parameter μ _a. The amplitude filter coefficient a _{(n + 1)} is updated by adding / subtracting the update term of the amplitude update formula. The adaptive filter coefficient updating unit 180 calculates the previously updated phase filter coefficient φ _{(n) based} on the replacement estimation transfer function Qh, the residual signal e _(n), and the phase step size parameter μ _φ. The phase filter coefficient φ _{(n + 1)} is updated by adding / subtracting the update term of the phase update formula.

(3.3) Overview of Calculation Update Processing of Gh Map Data, P Data, and Qh Map Data An overview of calculation update processing of Gh map data, P data, and Qh map data will be described. The calculation processing of Gh map data, P data, and Qh map data is performed at the initial stage of manufacture, and is performed every time the engine 10 is stopped, for example. That is, the Gh map data stored in the Gh map data storage unit 210, the negative feedback feedback element P stored in the P data storage unit 220, and the Qh map data stored in the Qh map data storage unit 230 are sequentially updated. .

  First, as shown in FIG. 9, Gh map data is calculated (step S1). Subsequently, based on the Gh map data, the negative feedback feedback element P is calculated using the pole placement method (step S2). Thereafter, Qh map data is calculated using the Gh map data and the negative feedback feedback element P.

(3.4) Calculation Update Processing of Gh Map Data and Negative Feedback Feedback Element P With reference to FIG. 10, functional blocks regarding calculation update processing of Gh map data and P data will be described. As shown in FIG. 10, in the active vibration noise suppression apparatus 100, the configuration used for the calculation update processing of Gh map data and P data includes a frequency setting unit 310, a filter coefficient setting unit 320, and a first control signal generation. Unit 120, third control signal generation unit 140, generator 150, residual signal detection unit 160, Gh map data calculation update unit 330, P data calculation update unit 340, Gh map data storage unit 210, A P data storage unit 220 is provided.

The frequency setting unit 310 sets the frequency f used for the first control signal y1 _(n) for calculating (identifying) the Gh map data. For example, the range of 30 Hz to 70 Hz in the frequency band of the engine 10 is set as the control target frequency. The filter coefficient setting unit 320 sets the amplitude filter coefficient a _(n) and the phase filter coefficient φ _(n) used for the first control signal y1 _(n) for calculating (identifying) the Gh map data.

The first control signal generation unit 120 uses the frequency f set by the frequency setting unit 310 and the filter coefficients a _(n) and φ _(n) set by the filter coefficient setting unit 320 based on the equation (13). ) To generate a first control signal y1 _(n) obtained. Here, the first control signal generation unit 120 is functionally the same as the first control signal generation unit 120 described above, and therefore is given the same reference numeral.

The third control signal generation unit 140 outputs the first control signal y1 _(n) generated by the first control signal generation unit 120 as the third control signal y3 _(n) . Here, the second control signal y2 _(n) is not used. Note that the third control signal generation unit 140 here is functionally identical to the above-described third control signal generation unit 140, and thus is given the same reference numeral.

The generator 150 is the same as the generator 150 described above, and generates a control vibration or control sound according to the third control signal y3 _(n) (= y1 _(n) ) output by the third control signal generator 140. appear. At the evaluation point 20, the control vibration generated by the generator 150 or the like is transmitted through the transmission system B as vibration or sound Z _(n) . This vibration or sound Z _(n) is expressed by equation (23).

As described above, the residual signal detection unit 160 is arranged at the evaluation point 20 and detects the vibration or sound e _(n) at the evaluation point 20. Here, the vibration or sound e _(n) detected by the residual signal detection unit 160 is equal to Z _(n) .

The Gh map data calculation / update unit 330 includes the first control signal y1 _(n) generated by the first control signal generation unit 120 and the vibration or sound e _(n) at the evaluation point 20 detected by the residual signal detection unit 160. Based on the above, the amplitude component Ah _G and the phase component Φh _G of the first estimated transfer function Gh are calculated. The Gh map data calculation update unit 330 newly calculates Gh map data already stored in the Gh map data storage unit 210 even after storing the Gh map data in the Gh map data storage unit 210 as an initial state. Update to Gh map data.

  The P data calculation update unit 340 calculates the negative feedback feedback element P by using the above-described pole placement method using the first estimated transfer function Gh calculated by the Gh map data calculation update unit 330. The P data calculation / update unit 340 updates the P data already stored in the P data storage unit 220 to the newly calculated P data even after the P data is stored in the P data storage unit 220 as an initial state. To do. The P data calculation update unit 340 updates the P data each time the Gh map data calculation update unit 330 updates the first estimated transfer function Gh.

  Thus, the Gh map data stored in the Gh map data storage unit 210 and the P data stored in the P data storage unit 220 are updated every time the engine 10 is stopped, for example. The data is adapted to the state of the real transfer function G. That is, even if the first actual transfer function G changes with time, the currently stored Gh map data and P data are updated following the change with time.

(3.5) Qh Map Data Calculation Update Processing Next, functional blocks for Qh map data calculation update processing will be described with reference to FIG. Here, the Qh map data can be calculated by calculation based on the Gh map data and the P data. That is, the Qh map data can be sequentially updated by calculation after the Gh map data and P data calculation update processing described above. However, since the first estimated transfer function Gh is used, the replacement estimated transfer function Qh may be further deviated from the actual replacement transfer function Q. Therefore, in this embodiment, the Qh map data is actually measured, and the replacement estimated transfer function Qh is directly calculated and updated.

  As shown in FIG. 11, in the active vibration noise suppression apparatus 100, the configuration used for the calculation update process of the Qh map data includes a frequency setting unit 310, a filter coefficient setting unit 320, a first control signal generation unit 120, and the like. A second control signal generation unit 130, a third control signal generation unit 140, a generation device 150, a residual signal detection unit 160, a Qh map data calculation update unit 350, and a Qh map data storage unit 230. Yes.

The frequency setting unit 310 sets the frequency f used for the first control signal y1 _(n) for calculating (identifying) the Qh map data. The filter coefficient setting unit 320 sets the amplitude filter coefficient a _(n) and the phase filter coefficient φ _(n) used for the first control signal y1 _(n) for calculating (identifying) the Qh map data. The first control signal generation unit 120 uses the frequency f set by the frequency setting unit 310 and the filter coefficients a _(n) and φ _(n) set by the filter coefficient setting unit 320 based on the equation (13). ) To generate a first control signal y1 _(n) obtained.

The second control signal generation unit 130 transmits the third control signal y3 _(n−1) generated by the third control signal generation unit 140 via the first estimated transfer function Gh and the negative feedback feedback element P. The second control signal y2 _(n) in the case where it is assumed is generated. Third control signal generating unit 140 outputs a third control signal _y3 from the first control signal _{y1 (n)} by subtracting the second control signal _{y2 (n) (n)} as a control signal.

The generator 150 is the same as the generator 150 described above, and controls according to the third control signal y3 _(n) (= y1 _(n) -y2 _(n) ) output by the third control signal generator 140. Generates vibration and control sound. At the evaluation point 20, the control vibration generated by the generator 150 or the like is transmitted through the transmission system B as vibration or sound Z _(n) .

As described above, the residual signal detection unit 160 is arranged at the evaluation point 20 and detects the vibration or sound e _(n) at the evaluation point 20. Here, the vibration or sound e _(n) detected by the residual signal detection unit 160 is equal to Z _(n) .

The Qh map data calculation / update unit 350 includes the first control signal y1 _(n) generated by the first control signal generation unit 120 and the vibration or sound e _(n) at the evaluation point 20 detected by the residual signal detection unit 160. Based on the above, the amplitude component Ah _Q and the phase component Φh _Q of the replacement estimation transfer function Qh are calculated. The Qh map data calculation update unit 350 newly calculates Qh map data already stored in the Qh map data storage unit 230 even after storing the Qh map data in the Qh map data storage unit 230 as an initial state. Update to Qh map data.

  Thus, since the Qh map data stored in the Qh map data storage unit 230 is updated every time the engine 10 is stopped, for example, the Qh map data is adapted to the current state of the first actual transfer function G, The data takes the second control signal generation unit 130 into consideration. That is, even if the first actual transfer function G changes over time, the currently stored Qh map data is updated following the change over time.

<4. Active vibration and noise suppression device when applied to map control>
Next, an active vibration noise suppression apparatus 200 when the control concept described with reference to FIG. 3 is applied to map control (feedforward control) will be described with reference to FIGS.

(4.1) Control Block Diagram First, a control block diagram of the active vibration noise suppression device 200 will be described with reference to FIGS. 12 and 13. The filter coefficient setting unit 420 constituting the active vibration noise suppression apparatus 200 includes an estimated value Qh of the replacement transfer function Q stored in the Qh map data storage unit 230 and vibration or noise generated by the vibration noise generation source. Using the frequency f, the amplitude filter coefficient a and the phase filter coefficient φ are calculated.

  Then, the first control signal generation unit 120 generates the first control signal y1 based on the adaptive filter coefficient W and the frequency f. The second control signal generation unit 130 generates the second control signal y2 when the third control signal y3 is transmitted via the first estimated transfer function Gh and the negative feedback feedback element P. Here, the first estimated transfer function Gh and the negative feedback feedback element P are sequentially calculated.

  The third control signal generation unit 140 generates a third control signal y3 obtained by subtracting the second control signal y2 from the first control signal y1. Then, the third control signal generation unit 140 outputs the third control signal y3 as a control signal. The output third control signal y3 is transmitted to the evaluation point 20 via the first actual transfer function G. At the evaluation point 20, the control vibration or control sound Z to which the third control signal y3 is transmitted via the first actual transfer function G and the vibration or noise generated by the vibration noise generating source 10 are the second actual transfer function H. The vibration or the noise X transmitted through the antenna interferes.

  Here, the region surrounded by the alternate long and short dash line in FIG. 12 corresponds to the replacement transfer function Q. That is, FIG. 12 can be expressed as shown in FIG. According to FIG. 13, at the evaluation point 20, the control vibration or control sound Z to which the first control signal y 1 is transmitted via the replacement transfer function Q and the vibration or noise generated by the vibration noise generation source 10 are second actual. The vibration or noise X transmitted via the transfer function H interferes.

That is, also in the map control, it can be considered that the transfer function of the first control system is replaced with the replacement transfer function Q from the first actual transfer function G. Even if the actual resonance frequency f _G0 of the first actual transfer function G is included in the control target frequency range, the resonance frequency f _Q0 of the replacement transfer function Q is obtained by using the pole placement method described above. It can be changed outside the range. That is, the phase [Phi _Q substituent transfer function Q does not vary significantly in the control frequency range of interest. Therefore, even if the phase Φ _{G of} the first actual transfer function G and the phase Φh _{G of} the first estimated transfer function Gh are greatly shifted, the phase Φ h of the phase Φ _Q of the replacement transfer function Q and the estimated value Qh of the replacement transfer function Q It is possible to suppress a large deviation from _Q. Therefore, vibration and noise at the evaluation point 20 can be reliably suppressed in the control target frequency range. In particular, even if the first transmission system changes due to changes over time or temperature, vibration and noise can be reliably suppressed.

(4.2) Functional Block of Entire Control A case where the above-described control block diagram is applied as an actual functional block will be described with reference to FIG. Here, as an actual functional block, an automobile will be described as an example as in the case of adaptive control.

  The active vibration noise suppression apparatus 200 includes a frequency calculation unit 110, a first control signal generation unit 410, a second control signal generation unit 130, a third control signal generation unit 140, a generation device 150, a filter coefficient A setting unit 420, a Gh map data storage unit 210, a P data storage unit 220, and a Qh map data storage unit 230 are provided. Here, the same components as those in the adaptive control shown in FIG. 14 described above are denoted by the same reference numerals, and detailed description thereof is omitted.

First control signal generation unit 410, frequency f calculated by frequency calculation unit 110, amplitude filter coefficient a _(n) and phase filter coefficient φ _(n) set by filter coefficient setting unit 420 described later Based on this, the first control signal y1 _(n) as a sine wave obtained according to the equation (23) is generated by map control. Here, the subscript _(n) is a subscript representing the sampling number (time step). That is, as apparent from the equation (23), the first control signal y1 _(n) includes the frequency f, the amplitude filter coefficient a _(n) as the adaptive filter coefficient W _(n) , and the phase filter coefficient φ _(n). Is a signal at time t _(n) .

From the map data of the replacement estimated transfer function Qh stored in the filter coefficient setting unit 420 and the Qh map data storage unit 230, the replacement estimated transfer function Qh corresponding to the frequency f calculated by the frequency calculating unit 110 is selected. To do. The replacement estimated transfer function Qh is stored in the map data of the replacement estimated transfer function Qh. The replacement estimated transfer function Qh is represented by an amplitude component Ah _Q and a phase component Φh _Q corresponding to the frequency f.

10: vibration noise generation source, 20: evaluation point, 100, 200: active vibration noise suppression device 110: frequency calculation unit, 120: first control signal generation unit, 130: second control signal generation unit 140: third control 150: generator, 160: residual signal detector 170: Qh data selection unit, 180: adaptive filter coefficient update unit 210: Gh map data storage unit, 220: P data storage unit 230: Qh map data storage unit 310: Frequency setting unit 320: Filter coefficient setting unit 330: Gh map data calculation update unit 340: P data calculation update unit 350: Qh map data calculation update unit 410: First control signal generation unit 420: Filter coefficient Setting section

Claims (5)

An active vibration noise suppression device that generates control vibration or control sound according to a control signal and actively suppresses vibration or noise at an evaluation point,
A frequency calculator that calculates the frequency of the source of vibration or noise;
A first estimated value Gh of the first actual transfer function G in the first transmission system up to the evaluation point is calculated via the control vibration control sound generator that generates the control vibration or control sound after outputting the control signal. A transfer function estimated value calculation unit;
When the transfer function of the first transfer system is replaced with a replacement transfer function Q, which is a state in which a negative feedback feedback element P is added to the first actual transfer function G, using the pole placement method, Based on the estimated value Gh of the transfer function G, a feedback element calculation unit that calculates the negative feedback feedback element P for replacing the actual resonance frequency f _G0 in the first transmission system with a frequency f _Q0 outside the control target frequency range. When,
A replacement transfer function estimated value calculation unit for calculating an estimated value Qh of the replacement transfer function Q based on the estimated value Gh of the first actual transfer function G and the negative feedback feedback element P;
A first control signal generator that generates a first control signal y1 based on the first filter coefficient calculated by the estimated value Qh of the replacement transfer function Q and the frequency calculated by the frequency calculator;
A second control signal generator for generating a second control signal y2 based on the first control signal y1, the negative feedback feedback element P and the frequency;
Generating a third control signal y3 by subtracting the second control signal y2 from the first control signal y1, and outputting the third control signal y3 as the control signal;
An active vibration noise suppression device comprising: In claim 1,
The replacement transfer function estimated value calculation unit outputs the third control signal y3 obtained by subtracting the second control signal y2 from the first control signal y1 as the control signal when the third control signal generation unit outputs the control signal. An active vibration noise suppression device that calculates an estimated value Qh of the replacement transfer function Q based on vibration or sound at the evaluation point according to the third control signal y3. In claim 1 or 2,
The first transfer function estimated value calculation unit applies vibration or sound at the evaluation point according to the first control signal y1 when the third control signal generation unit outputs the first control signal y1 as the control signal. Updating the estimated value Gh of the first actual transfer function G based on
The feedback element calculation unit is an active vibration noise suppression apparatus that updates the negative feedback feedback element P based on the updated estimated value Gh of the first actual transfer function G. In any one of Claims 1-3,
The active vibration noise suppression device further includes a residual signal detection unit that detects a residual signal due to interference between vibration or noise caused by the generation source and the control vibration or control sound at the evaluation point,
The first control signal generation unit includes a first filter coefficient calculated by an estimated value Qh of the replacement transfer function Q, the frequency calculated by the frequency calculation unit, and the residual signal detection unit. An active vibration noise suppression device that generates the first control signal y1 by an adaptive control method based on a residual signal. In any one of Claims 1-3,
The first control signal generation unit is configured to perform the first control signal by the feedforward control method based on the first filter coefficient calculated by the estimated value Qh of the replacement transfer function Q and the frequency calculated by the frequency calculation unit. An active vibration noise suppression device that generates a control signal y1.

Images