JP3506285B2 - Adaptive control method for periodic signals - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of adaptively controlling a periodic signal, and more specifically, it adds a signal generated adaptively to a periodic signal to be removed so that a specific frequency of the periodic signal can be eliminated. The present invention relates to an adaptive control method that actively removes the influence of components. Here, since the periodic signals to be removed are various such as vibration, noise, electromagnetic waves, and electric noise, the control method of the present invention can be applied to an extremely wide industrial field.

[0002]

2. Description of the Related Art Regarding an adaptive control method for periodic signals,
There are already several control theories and their application examples as so-called active control. For example, in the feature article of Vol. 32, No. 4 (April 1993) of "Measurement and Control" published by the Society of Instrument and Control Engineers. Is also introduced.

(FX-LMS) Conventionally, an adaptive digital filter technology has been used as an active cancellation system for various noises and vibrations, and in particular, a Filtered-X LMS algorithm (abbreviated as FX-LMS) schematically shown in FIG. 21 is used. Widely used. In addition, as a modification of this, as schematically shown in FIG. 22, an FX-LMS in which the estimated value of the transfer characteristic G of the controlled object is taken into consideration.
There is also a control method.

However, in the FX-LMS, a convolution operation is required when generating the reference signal, and a large number of taps that differ depending on the sampling period are required to properly realize the impulse response of the system. . Therefore, the amount of processing data becomes enormous, and the convolution calculation for the number of taps is also required for the calculation of the filter coefficient accompanying this, so the calculation amount is still increased. In particular, when dealing with a plurality of input / output signals, such an increase in the amount of calculation becomes more remarkable, and not only the capacity of the calculation device cannot keep up, but also
There is a possibility that proper convergence characteristics of the filter coefficient may not be obtained.

(SFX) This problem is solved by eliminating FX
In order to reduce the calculation amount of LMS, the synchronous adaptive algorithm (Synchronize) schematically shown in FIG.
d Filtered-X Algorithm, abbreviation S
FX) was developed. SFX is intended for periodic signals or quasi-periodic signals, and an impulse train synchronized with the fundamental period of a periodic input signal is generated inside the processor so that FX-LMS can be applied with this as an imaginary input. It was done. The SFX algorithm is specifically described in the section of the prior art of Japanese Patent Application No. 6-2013384 "Adaptive control method of periodic signal". SFX
Since the convolution calculation is not necessary and the calculation amount can be reduced by using, the sampling cycle can be set faster and the control capability can be improved.

[0006]

However, the above S
In the FX, when the frequency of the periodic signal to be controlled increases, the sampling period becomes shorter in synchronization with this, so that the tap number of the impulse response needs to be increased accordingly. As a result, due to the increase in the calculation time required for these processes and the shortening of the sampling cycle, there are still problems that the capacity of the calculation device is insufficient, and that the calculation accuracy must be reduced to compensate for the shortage. It was

(DXHS: Prior Art) An adaptation of a periodic signal which solves this problem and can improve the signal removal characteristic of the system without causing a decrease in calculation accuracy due to the shortening of the sampling period of the periodic signal. Regarding the control method, the applicant of the present application has already applied for the above-mentioned Japanese Patent Application No. 6-201384. The prior art control method described in that specification is now described in Delayed-X Harmo.
It will be referred to as a “nics Synthesizer algorithm” (abbreviated as DXHS). The outline of the control method is
A block diagram is schematically shown in FIG.

The DXHS is intended to suppress the fundamental wave of a signal having periodicity and its harmonics, such as noise and vibration generated by rotation of an automobile engine, rotation of a propeller of an airplane, and rotation of a rotor of a helicopter. , An adaptive control method for removing the specific frequency component. That is, DXHS obtains a gradient vector by partially differentiating an evaluation function represented by the square of a function including a sine wave output signal by a filter coefficient W that is a function of the amplitude and phase of the output signal, and the gradient vector is constant. By subtracting a product of the numbers from the filter coefficient, the filter coefficient is updated every time, and the amplitude and phase of the sine wave output signal are updated with the updated filter coefficient amplitude and phase. is there. With this calculation method, DXHS eliminates the convolution operation that was conventionally required for output calculation,
It is possible to reduce the calculation amount and prevent deterioration of calculation accuracy.

(Problems of Prior Art and Prior Art) By the way, in each of the above-mentioned conventional art and prior art algorithms, the transfer characteristic including the delay element of the system is usually the impulse response, or the gain and phase delay at each frequency. For example, the filter coefficient is updated based on these data.

However, when applied to an actual system, the transfer characteristic of the system does not always remain in the state measured at the initial stage. For example, when targeting a system such as an automobile, there is a difference in transmission characteristics between when the vehicle is stopped and when the vehicle is running, and it depends on weather conditions such as temperature and the loading state of passengers. The transfer characteristics are different even after traveling 10,000 km. As described above, in a system that is an actual control target, there is always a possibility that its transfer characteristic changes. With respect to such a change in the transfer characteristic, in particular, a change in the delay characteristic represented by the phase delay, the above-mentioned conventional technology and the prior art can only cope with a very limited range.

This problem can be concretely understood by comparing FX-LMS and DXHS as examples. The experiment was conducted using the electric circuit shown in FIG.
The phase conversion amplifier in the circuit is for artificially creating a phase delay, that is, a change in the delay characteristic, and for checking the followability of both algorithms with respect to the change in the phase delay characteristic of the controlled system.

As a result, the FX-LMS is shown in FIG.
For 6 and DXHS, the convergence characteristics shown in FIG. 27 were obtained. That is, both FX-LMS and DXHS have a phase change ±
If it is about 60 degrees, the response can be dealt with and the response will converge, but if it reaches ± 90 degrees, it will not be able to follow and the response of the system will diverge. Therefore, according to not only the conventional technology but also the prior art, when the phase delay of the transfer characteristic of the system to be controlled deviates greatly from a preset value, it becomes impossible to stably control the periodic signal to be suppressed. There was a point.

(Problem of the present invention) Therefore, the present invention responds to changes in the transfer characteristics (gain and phase delay at each frequency) of the system to be controlled, and in particular to phase changes, that is, large changes in delay characteristics over time. Then, following this, it is an object to be solved to provide an adaptive control method of a periodic signal which can suppress the influence of the periodic signal and converge the response of the system.

[0014]

SUMMARY OF THE INVENTION An adaptive control method for a periodic signal according to the present invention which solves the above-mentioned problems, is a primary basic sine signal synchronized with a periodic signal which affects an observation point. Wave and / or an adaptive signal composed of the fundamental sine wave to the M-th order (2 ≦ M) harmonic signals of the fundamental sine wave in antiphase to add the specific frequency component of the periodic signal to the observation point. An adaptive control method for a periodic signal that actively removes the influence of the above, wherein the adaptive signal generation algorithm generates the adaptive signal y (n) based on the primary angular frequency of the periodic signal at time n, As shown in Equation 4,
Each order of the adaptive signal y (n) (order k = 1, 2, ...,
M) sine wave amplitude a _k and phase φ _k and phase delay
Adaptive engagement to component convergence stability factor G _{k [p]} hat about
The number vector W (n) should be removed to eliminate the effect of the periodic signal.
Phase based on the error signal e (n) detected at the observation point
According to the equation 5 including the adjustment parameter ψ, every time the time n elapses
To the adaptive coefficient vector W for the amplitude and phase of the periodic signal and the transfer characteristic of the controlled system.
(N) adaptive coefficient vector updating algorithm that adaptively adjusts each of the components, and an amplitude a _k and a phase φ of each order that is a part of the components of the updated adaptive coefficient vector.
_{With k} , the amplitude a _k and the phase φ _k of the sine wave of each order of the adaptive signal y (n) are updated.

[0015]

[Formula 4]

[0016]

[Formula 5]

In the adaptive coefficient vector updating algorithm of the above-mentioned embodiment , the phase adjustment parameter is updated according to the equation 6 equivalent to the equation 5 in which ψ = π / 2.

[0018]

[Formula 6]

In the adaptive control method described above, equivalent transfer characteristic data that is an estimated value of a phase delay corresponding to each angular frequency of the transfer characteristic from the adaptive signal to the observation point is included, and the adaptive coefficient vector The convergence stability coefficient G _{k [p]} hat corresponding to each order that is a component of W (n)
, Said each of the primary angular frequency is varied, is the initial value can also algorithms configuration provided from corresponding the equivalent transmission characteristic data on angular frequency of each order.

Here, the equivalent transfer characteristic data may be set in advance before the start of adaptive control. Further, the equivalent transfer characteristic data may be updated by the convergence stability coefficient G _{k [p]} hat in the updated component of the adaptive coefficient vector W (n). Also, the step size parameter is
An algorithm configuration in which the numerical value is changed according to the value of the primary angular frequency is also possible.

[0021] Incidentally, the adaptive coefficient vector update algorithm determines the gradient vector by partially differentiating the square of the error signal e (n) in the previous term adaptive coefficient vector W (n), respectively, each component of the gradient vector The adaptive coefficient vector W (n) multiplied by the step size parameter of is subtracted from the adaptive coefficient vector W (n) to calculate the adaptive coefficient vector updated every time the time n elapses. is there.

Further, the adaptive coefficient vector updating algorithm, it is also possible to take the following three forms.
That is, as another form , in the adaptive coefficient vector updating algorithm, the adaptive coefficient vector W (n) is used.
As shown in Equation 7 , the error signal e is composed of the amplitude a _k and the phase φ _k of the sine wave of each order of the adaptive signal y (n) and the convergence stability coefficient G _{k [g]} hat relating to the gain as a component. It is updated according to equation ( 8 ) based on (n).

[0023]

[Formula 7]

[0024]

[Formula 8]

As another form, in the adaptive coefficient vector updating algorithm, the adaptive coefficient vector W (n) is expressed by the adaptive signal y as shown in Equation 9.
The amplitude a _k and the phase φ _k of the sine wave of each order of (n) and the convergence stability coefficient G _{k [p]} hat relating to the phase delay are used as components, and are updated according to Formula 10 based on the error signal e (n). Is.

[0026]

[Formula 9]

[0027]

[Formula 10]

As another form, in the adaptive coefficient vector updating algorithm, the adaptive coefficient vector W (n) is expressed by the adaptive signal y as shown in Expression 11.
The error signal e is composed of the amplitude a _k , the phase φ _k, and the convergence stability coefficient G _{k [g]} hat relating to the gain and the convergence stability coefficient G _{k [p]} hat relating to the phase lag of the components _(n) as components. It is updated according to the equation ( 12 ) based on (n).

[0029]

[Formula 11]

[0030]

[Equation 12]

[0031]

By the way, the step size parameter is
The 3M or 4M components of the adaptive coefficient vector W (n) can be adjusted or set independently. Further, it is easy to set the update cycle (the same as the normal sampling cycle) of the adaptive coefficient vector W (n) and the step size parameter to a positive constant, but they may be variable depending on the system state or the passage of time. . Further, the algorithm for updating the adaptive coefficient vector W (n) can also be switched according to the state of the system.

[0033]

In the adaptive control method for periodic signals of the present invention, each order (order k = 1, 2, ..., M) of the adaptive signal y (n) is used as a component of the adaptive coefficient vector W (n). Not only the amplitude a _k and the phase φ _k of the sine wave of, but also the convergence stability coefficient G _k hat (1 ≦ k ≦ M) corresponding to the transfer characteristic of the controlled system.

This convergence stability coefficient G _k hat is composed of at least one of a convergence stability coefficient G _kg hat relating to gain and a convergence stability coefficient G _kp hat relating to phase delay, and the variation of the transfer characteristic of the controlled system during the adaptation process. It has the effect of absorbing and helping to adapt. That is, when the adaptive coefficient vector W (n) is updated by the above-described update algorithm based on the error signal observed at the observation point where the influence of the periodic signal should be removed, the following action occurs. . First, the amplitude and phase of each order of the adaptive signal are adjusted so as to reduce the square of the error signal. At the same time, the variation in the transfer characteristic of the controlled system can be adjusted by adjusting the convergence stability coefficient.

Therefore, by having the convergence stability coefficient G _k hat in the component of the adaptive coefficient vector W (n), it becomes possible to suppress the error signal due to the fluctuation of the transfer characteristic of the controlled system. Here, the adaptive control algorithm can be configured by the least squares algorithm by using a gradient vector introduced as the updating algorithm of the adaptive coefficient vector W (n). That is, the adaptive coefficient vector W (n) shown in the equations (4), (6), and (8) and the update algorithm corresponding to each of them ((5)
The adaptive control algorithm can be configured by the least-squares algorithm by adopting any one of the combinations with 7, 9), and the influence of the periodic signal to be suppressed can be suppressed.

If the adaptive coefficient vector W (n) shown in the above equation 10 (equivalent to the above equation 6) and the corresponding update algorithm (the above equation 11 or equation 12) are adopted, the above gradient vector is obtained. Better convergence than the algorithm using. In particular, when the update algorithm shown in Expression 12 is used, a smaller amount of calculation is required, and
Good control results can also be achieved.

Further, it has an equivalent transfer characteristic data, and an estimated value of the transfer characteristic of the controlled system suitable for each frequency is
If the convergence stability coefficient G _k in the adaptive coefficient vector W (n) is given as an initial value, the convergence of the adaptive coefficient vector W (n) can be accelerated. Further, if the equivalent transfer characteristic data is set to an appropriate value in advance by the vibration test or the numerical analysis before the adaptive control is started, a better initial value is given, so that the convergence becomes faster. Then, if the equivalent transfer characteristic data is updated by the convergence stability coefficient, even if the transfer characteristic of the controlled system varies greatly, it is possible to always prepare the equivalent transfer characteristic data which gives the initial value adapted to it.

By the way, the step size parameter, the parameter such as the update period, and the adaptive coefficient vector W
If either (n) or its update algorithm is appropriately switched or adjusted depending on the state of the system including the primary angular frequency, the adaptability to the fluctuation of the periodic signal and the controlled system is further improved. .

[0039]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The adaptive control method for periodic signals according to the present invention and four embodiments in which the method is applied to a physical system are shown in FIG.
~ It demonstrates based on FIG. [Description of Adaptive Control Method of Periodic Signal of the Present Invention] The adaptive control method of the periodic signal of the present invention and the development process thereof will be described below with reference to FIGS. 1 and 2.

(Theoretical Development) First, the theoretical development corresponding to the first half of the development process of the adaptive control method of the periodic signal of the present invention will be described with reference to FIG. According to the adaptive control method of the periodic signal of the present invention, with respect to the periodic signal d (n) whose effect should be removed, the adaptive signal y (n) consisting only of the primary fundamental sine wave synchronized with this periodic signal. Or
The adaptive sine wave y (n) consisting of the fundamental sine wave and its harmonic signals up to the Mth order (2 ≦ M) is actively generated. This adaptive signal y (n) is transmitted to become an adaptive transmission signal z (n), which is in anti-phase with the periodic signal d.
The periodic signal d (n) is added to (n).
The fundamental wave component or the specific frequency components from the 1st order to the Mth order are canceled.

Therefore, the adaptive control method of the periodic signal of the present invention does not remove the periodic signal d (n) itself, but cancels the influence of the same signal at the observation point 24 affected by the same signal. , The control object is to converge the error signal e (n) to zero. In addition, "first-order basic sine wave"
Is the "fundamental wave" or "first harmonic" or "fundamental vibration",
The "○ order harmonic" is sometimes called "○ order harmonic" or "○ order higher vibration".

The overall configuration of the control system of the present invention is shown in FIG.
As shown in FIG. 3, the adaptive control algorithm 1 for exchanging signals with each other and the physical system 2 to be controlled are included. The adaptive control algorithm 1 is based on an update algorithm 11 of an adaptive coefficient vector W (n) driven by an error signal e (n), which will be described later, and elements of the updated adaptive coefficient vector W (n), from 1st to Mth order. The adaptive signal y
An adaptive signal generation algorithm 12 for generating (n),
It consists of the equivalent transfer characteristic data 13 (G _{g, p} hat) of the controlled system G that gives an initial value to the convergence stability coefficient G _k hat of the adaptive coefficient vector W (n). The definition and operation of each element of the adaptive control algorithm 1 will be described in detail after the physical system 2 to be controlled is described. (Here, "hat" is a notation indicating that it is an estimated value, and in each figure and each mathematical expression , "G" is attached with a mountain-shaped symbol (also referred to as a roof). .. On the other hand, the physical system 2 to be controlled has the periodic signal generation system 22 for generating the periodic signal d (n) whose effect should be removed as an uncontrolled (or insufficiently controlled) unique system. is doing. The periodic signal generation system 22 includes a signal generation source 20 that is a generation source of a periodic signal and a signal transmission characteristic 21 (G ′ _{g, p} ) that transmits the signal to an observation point 24. (Here,
About _{G'g, p} in the text of the specification,
In the formula, the brackets [] on both sides of the subscripts g and p are digitized.
It should be added that it is omitted in the process of application.
For subscripts g and p, omit the same parentheses [].
Abbreviations can be found in various places in the text of the specification, but read in correspondence with symbols.
There is no room for misunderstanding by the user, so leave it as it is.
It )

In many cases, the periodic signal d (n) generated by such a periodic signal generation system 22 is expressed by the following equation 13 as a combination of the primary fundamental wave and its harmonics. Can be expressed. That is, as a periodic signal obtained by synthesizing a k-th order (1 ≦ k ≦ L) sine wave having a primary angular frequency (also referred to as a fundamental angular frequency) of ω ^* and amplitude and phase of a ^* _k and φ ^* _k , The periodic signal d (n) measured at the observation point 24 is formulated. This is a formulation based on Fourier decomposition (harmonic analysis) that decomposes a periodic function.

[0044]

[Equation 13]

The periodic signal d (n) to be canceled is the adaptive signal y (n) input from the adaptive control algorithm 1.
Is combined with the adaptive transfer signal z (n) transferred via the transfer characteristic 23 (G _{g, p} ) of the controlled system to produce an error signal e (n) measured at the observation point 24. Therefore, an appropriate adaptive signal y (n) is input to the physical system 2 to be controlled, and an adaptive transfer signal z (n) having the same amplitude and opposite phase as the specific component to be erased of the periodic signal d (n) is generated. If it is possible, the same component can be canceled out, and the error signal e (n) can be suppressed sufficiently low.

For this purpose, the adaptive control algorithm 1 is designed. An adaptive signal generation algorithm 12 for generating an appropriate adaptive signal y (n) has a first-order angular frequency of ω, an amplitude and a phase of a _k , φ _k, and an M-th order from a first-order fundamental wave shown in the following Expression 14. The adaptive signal y (n) is generated as a periodic signal obtained by synthesizing the harmonics up to. However,
In some cases, M = 1 and the adaptive signal y (n) is composed of only the fundamental wave.

[0047]

[Equation 14]

Here, the order M of the adaptive signal y (n) is usually set to the highest order of harmonics to be suppressed, which is less than or equal to the order L of the periodic signal d (n) whose influence should be removed. Strictly speaking, it can be said that the order L is usually infinite in the actual physical system, and it is wise to reduce the order M of the adaptive signal to the minimum necessary in accordance with the vibration mode to be removed and to reduce the cost of the control system. . Of the periodic signals d (n) of order L, higher-order signals (also called spillover) exceeding the order M of the appropriately set adaptive signal y (n) often have small amplitude and attenuation. Being good, it is rarely a problem in practice.

As described above, the adaptive signal y (n) whose order M is properly set is the primary angular frequency ω, and the amplitude a _k and phase φ of the sine wave (angular frequency kω) of each order. _k
It is uniquely defined by defining (k = 1, 2, ..., M). Of these, the primary angular frequency ω is obtained by directly measuring the true value ω ^* of the primary angular frequency from the signal generation source 20 of the physical system 2 to be controlled. In most cases, this measurement can be performed precisely, and the true value ω ^* of the primary angular frequency and the measured value ω are treated as engineering equivalent (ω ^* = ω). The primary angular frequency ω is the periodic signal d (n) as an alternative when the measurement from the signal source 20 is difficult.
You may ask from

On the other hand, the amplitude a _k and the phase φ _k of the _kth sine wave of the adaptive signal y (n) are obtained as the elements of the adaptive coefficient vector W (n) updated at each step. That is, the adaptive coefficient vector W (n) is composed of the amplitude a _k and phase φ _k of the sine wave of each order and the convergence stability coefficient G _k hat (1 ≦ k ≦ M) as shown in the following Expression 15. Is defined as

[0051]

[Equation 15]

The introduction of the convergence stability coefficient G _k hat into the elements of the adaptive coefficient vector W (n) is a feature of the adaptive control method of the periodic signal of the present invention. The convergence stability coefficient G _k hat is composed of at least one of a convergence stability coefficient G _kg hat regarding gain and a convergence stability coefficient G _kp hat regarding phase delay. Both the convergence stability coefficients G _kg hat and G _kp hat of each order relate to the gain G _kg and the phase delay G _kp at each frequency kω of the transfer characteristic 23 (G _{g, p} ) of the controlled system. That is, the adaptive transfer signal z (n) obtained by transferring the adaptive signal y (n) shown in the above _equation 14 by the transfer characteristic G _{g, p} of the controlled system 23 is described as shown in the following equation 16. To be done.

[0053]

[Equation 16]

Then, as is apparent from FIG. 1, the error signal e (n) is the periodic signal d (n) and the adaptive transmission signal z.
It is defined as the arithmetic sum of (n). That is, e (n) = d (n) + z (n). Based on the error signal e (n), an updating algorithm 11 for the adaptive coefficient vector W (n) described later can be determined. Then, according to the algorithm 11, the convergence stability coefficient G _k among the components of the adaptive coefficient vector W (n)
The hat (the component is at least one of the G _kg hat and the G _kp hat) is the transfer characteristic G _{g, p} (at least one of the gain G _kg and the phase delay G _kp ) of the controlled system at each angular frequency kω. Respond to fluctuations. Here, G _kg and G _kp are some of the coefficients forming the error signal e (n), where the system is stable.

Now, the error signal e defined as above
When the square of (n) is partially differentiated by the adaptive coefficient vector W (n), a gradient vector ∇ (n) is obtained as shown in the following Expression 17. However, here, the convergence stability coefficient G _k hat in the adaptive coefficient vector W (n) is assumed to be composed of both the convergence stability coefficient G _kg hat regarding the gain and the convergence stability coefficient G _kp hat regarding the phase delay.

[0056]

[Equation 17]

This gradient vector ▽ (n) is the error signal e
It suggests a direction to increase the expected value of the square of (n). Therefore, the adaptive coefficient vector W (n) is obtained by multiplying each component of the gradient vector ▽ (n) by an appropriate step size parameter.
By subtracting from, the adaptive coefficient vector W (n) can be appropriately converged. The step size parameter is
It can be made variable according to the state of the system,
In the following embodiments, a suitable positive constant is used.

Here, the gradient vector ∇ (n) is the four types of coefficients a _k , φ _k , G of the adaptive coefficient vector W (n).
_Since there is a component for each of _kg hat and G _kp hat, the step size parameter is accordingly
Prepare four types of μ _a , μ _p , μ _Gg , and μ _Gp . The step size parameter can be set independently for each component by adding a subscript k indicating the order, but here, for simplification, the step size parameters common to each of the above four types are set. Then, the update algorithm of the adaptive coefficient vector W (n) can be set as shown in the following Expression 18.

[0059]

[Equation 18]

If the step size parameter is set to an appropriate value, the updating algorithm 11 shown in the above equation 18 converges, and the amplitude a _k and the phase φ of the sine wave of each order of the adaptive signal y (n). _k can be set appropriately. As a result, the adaptive control algorithm 1 generates an adaptive signal y (n) that yields an adaptive transfer signal z (n) that cancels out the specific component of the periodic signal d (n) to be controlled, at the observation point 24. The error signal e (n) can be suppressed small. Thus, according to the adaptive control method for periodic signals of the present invention, the signal transfer characteristic G of the physical system 2 to be controlled is possessed.
_{g, p, G 'g,} it is possible to control the system to adapt to changes by _p.

By the way, prior to calculating the update algorithm 11 for the adaptive coefficient vector W (n), the equivalent transfer characteristic data 13 is given the primary angular frequency ω from the signal source 20 as described above. Then, in the equivalent transfer characteristic data 13, the G _kg hat regarding the gain and the G _kp hat regarding the phase delay at each frequency kω are used as initial values in the updating algorithm 11 of the adaptive coefficient vector W (n) based on the latest data. give.

Here, the latest data of the equivalent transfer characteristic data 13 refers to preset data for the angular frequency that the adaptive control algorithm 1 first experiences. Then, for the angular frequency experienced from the second time onward, it refers to the data updated by the convergence stability coefficient of the adaptive coefficient vector W (n). This operation can be newly performed corresponding to the fluctuation of the primary angular frequency ω.

Further, the above-mentioned preset data is
In this embodiment, prior to the adaptive control test, the transfer characteristic G _{g, p} of the controlled system 23 measured by performing a vibration test in which the frequency of the input sine wave is swept in the applicable range to check the gain and the phase delay is measured. The data. In addition to the vibration test, modal analysis or numerical calculation using a mathematical model can be used to obtain this data.

Therefore, based on this data (G _kg hat and G _kp hat), a Bode diagram showing an estimated value of the transfer characteristic G _{g, p} of the controlled system 23 can be created. The data in the equivalent transfer characteristic data 13 is the update algorithm 11 for the adaptive coefficient vector W (n).
Since the G _kg hat and G _kp hat are updated successively by this, it is always updated by using this. This update can be performed not only every time the update is performed but also as needed. Therefore, in Examples 1 and 2 described later, the previously recorded data is used only for the initial value of the adaptive coefficient vector W (n) at the first angular frequency ω.

(Narrowing down the object) In the above control method, the transfer characteristic G _{g, p} of the controlled system 23 is
This is an algorithm for adapting when both the gain G _kg and the phase delay G _kp greatly change. Therefore, for example, when the gain does not fluctuate significantly, the estimation of the gain G _kg can be stopped and the algorithm can be simplified to reduce the cost (calculation amount) of the control system (controller).

Therefore, in each of the embodiments described later _, the phase delay G in the transfer characteristic G _{g, p} of the controlled system 23 is set.
_With respect to _kp , the goal is to control the periodic signal by adapting to large changes. Therefore, as shown in the following equation 19, the convergence stability coefficient to be introduced into the adaptive coefficient vector W (n) is defined only for the phase delay (G _kp hat), and the adaptive coefficient vector W (n) is defined again.

[0067]

[Formula 19]

Therefore, the square of the error signal e (n) is
Partial differentiation with the newly defined adaptive coefficient vector W (n) yields a new gradient vector ∇ (n) consisting of three types of components, as shown in the following Expression 20.

[0069]

[Equation 20]

This gradient vector ∇ (n) has three kinds of coefficients a _k and a _k for each degree of the adaptive coefficient vector W (n).
Since there is a component for each of φ _k and G _kp hat, the step size parameters are correspondingly μ _a , μ _p , and μ
Prepare 3 types of _Gp . Then, the adaptive coefficient vector W
The update algorithm of (n) can be newly defined as shown in the following Expression 21.

[0071]

[Equation 21]

If the step size parameter is set to an appropriate value, the update algorithm 11 shown in the above equation 21 converges, as in the above algorithm. As a result, the adaptive control algorithm 1 can control the system by adapting to a large change in the phase delay of the signal transfer characteristics G _{g, p} and G ′ _{g, p} of the controlled physical system 2. .

In this case, as for the data stored in the equivalent transfer characteristic data 13 of the controlled system G, the estimated value G _kp hat regarding the phase delay is sufficient, and the estimated value G _kg hat regarding the gain is unnecessary. . (Trial Study) As described above, the adaptive control method of the periodic signal was derived from the system configuration shown in FIG.

[0074] However, in the updating algorithm of the number 21, is required convergence stability factor G _kg hat about gain, G _kg hat equivalent transfer characteristic data 13 is also disadvantageous occurs needs to be stored. Therefore,
The inventors have developed an update equation shown in the following Equation 22 excluding the G _kg hat and confirmed that it can work by numerical simulation. According to this update formula, it is not necessary to store the G _kg hat related to the gain in the equivalent transfer characteristic data 13.

[0075]

[Equation 22]

However, in the numerical simulation by the updating equation of the above formula 22, if the expected phase delay error is large, the convergence cannot be said to be sufficiently satisfactory, and there remains dissatisfaction for practical use. It was That is, when a mathematical model equivalent to Example 1 described later is used and numerical simulation is performed for a case where the difference between the phase data G _kp hat prepared in advance in the adaptive control system and the phase delay of the controlled system is large, sufficient convergence is achieved. I found that I could not get the sex. Specifically, it was difficult to converge when the difference was 170 degrees (50 Hz) and when the difference was 190 degrees (60 Hz). The step size parameter used at this time is μ _a = 1. , Μ _p = 10. , Μ _Gp =
1. Met.

As a result of consideration of the above phenomenon by the inventors, the reason why the update equation of the above equation 22 cannot obtain sufficient convergence is that the second component and the third component of the update equation have step size parameters. Except that they were the same. That is, therefore, it was presumed that the function of the convergence stability coefficient G _kp hat regarding the phase delay was not sufficiently exerted.

Therefore, a renewal formula shown in the following equation 23 in which the third component has a phase difference of −π / 2 with respect to the second component is experimentally proposed.

[0079]

[Equation 23]

When the updating equation of the above equation (23) was evaluated by the same numerical simulation as described above, a slight improvement in convergence was observed. That is, when a numerical simulation is performed for a case where the difference (fluctuation) in phase delay is large, the error signal e within about 0.1 second in the case of 170 degrees (50 Hz).
(N) could be converged. However, it was difficult to converge when they differ by 190 degrees (60 Hz).

Then, as a result of further trials conducted by the inventors, an updating algorithm 11 for the adaptive coefficient vector W (n) is obtained.
With respect to the above, it was possible to develop one having a better convergence than the one shown in the above equation 23. From the third component relating to the phase delay, the adaptive signal y
The amplitude a _k of each order in (n) is excluded.

[0082]

[Equation 24]

In the updating equation of the above equation 24, the third component relating to the phase delay is not a component for adjusting the gain, but a component intended to correspond to a large phase delay in the transfer characteristic of the controlled system. It was invented based on the idea. When this update formula was evaluated by the numerical simulation described above, the phase delay was 170 degrees (50H
z) Good convergence was obtained for both different cases and different cases of 190 degrees (60 Hz). Error signal e
The time required for convergence of (n) was 0.2 seconds for the former and about 0.05 seconds for the latter.

Therefore, it was concluded that the method of excluding the amplitude a _k from the third component and giving the phase difference from the second component as in the updating equation of the above equation 24 is effective. By the way, although the phase difference is limited to −π / 2 in the equation 24, there is no reason to be limited to this. Therefore, the inventors have introduced the phase adjustment parameter ψ, replaced the phase difference with −ψ, and developed the update equation shown in the following Expression 25.

[0085]

[Equation 25]

With respect to the update formula of this equation 25, ψ = 0, π
Numerical simulations of / 6, π / 3, ..., 11π / 6 at 30 ° intervals were conducted to evaluate the convergence.
The step size parameter at that time was the same as in the case of the numerical simulation described above. As a result, depending on the range of ψ, it was divided into a case where it converges and a case where it does not converge. That is, even when the phase delay of the controlled system deviates greatly from the expected phase, as shown in FIG.
The error signal e (n) in a certain range of ψ excluding the π part
Was able to converge. This area is considered to be fairly large.

The reason why the convergence deteriorates in the portion where ψ is ± π (that is, when it is an integer multiple of π) is that the second component and the third component of the updating equation of the above equation 25 have the same phase or opposite phase. The inventors suspect that they have been completely synchronized with. That is, the third component for updating the convergence stability coefficient G _kp hat related to the phase delay is added for the purpose of adapting to a large variation in the phase delay that cannot be applied only by updating the first component and the second component. Has been done. Therefore, if the second component and the third component of the updating equation are completely synchronized, it is considered that the adaptability to the fluctuation of the phase delay of the third component is impaired.

However, there is a possibility that the response can be converged even when ψ = 0 or ψ = ± π by appropriately setting the step size parameter and the like. However, since it takes a long time to converge and the stability is delicate, it is considered unsuitable for practical use. (Update Formula Applied to Example) Since the research results as described above have been obtained, the inventors have calculated that ψ = π /
The above-mentioned formula 24, which is limited to 2, is applied to the embodiment. In the formula 24, the exponential function expression is changed to the trigonometric function expression to obtain an equivalent update expression shown in the following expression 26.

[0089]

[Equation 26]

According to this update formula, it is possible to control with a smaller amount of calculation than that of the exponential function expression, and it is possible to reduce the cost of the control device. Further, similarly to this, the above-mentioned formula 14 for generating the adaptive signal y (n) is also
The same effect can be obtained by rewriting the trigonometric function expression. As described in detail above, by adopting this updating formula in the updating algorithm 11, the influence of the periodic signal d (n) is canceled more quickly,
The error signal e (n) can be converged. Therefore, in the following test of the present embodiment, the update algorithm 11 for the adaptive coefficient vector W (n) is configured by the adaptive coefficient vector W (n) shown in the above-mentioned Expression 19 and the update equation shown in the above Expression 26. ing.

[Example 1: Test and evaluation in electric circuit] Hereinafter, this example will be described with reference to FIGS. 1 and 3 to 7. As Example 1 of the present invention, the adaptive control method, which is an example of the present invention, was applied to an electric circuit including a function generator and a phase conversion amplifier, tested, and the control result was evaluated. The purpose of this example is to demonstrate how far the adaptive control method for periodic signals of the present invention can adaptively control following a large change in the phase delay, which is one of the transfer characteristics of the controlled system. Especially.

(Control Method of First Embodiment) As shown in FIG. 1, the adaptive control method of the periodic signal according to the present embodiment is implemented on a control system including the adaptive control algorithm 1 and the physical system 2 to be controlled. Has been done. Physical system 2
, A function generator is used to generate only the basic sine wave as the periodic signal d (n).

Therefore, the adaptive control algorithm 1 also
M = 1 is set and only the basic sine wave is supported. Here, the adaptive control algorithm 1 uses the adaptive coefficient vector W
(N) The update algorithm 11, the adaptive signal generation algorithm 12, and the equivalent transfer characteristic data 13 of the controlled system G are used. As the adaptive coefficient vector W (n), there are three convergence stability coefficients G _p hat in relation to the amplitude and phase of the adaptive signal y (n) of only the first-order fundamental sine wave (M = 1) and the phase delay in the above-mentioned mathematical expression 19. It uses a vector of components.

As the adaptive signal generation algorithm 12,
In the above formula 26, the one that updates the three components of only k = 1 is used. The initial value of the convergence stability coefficient G _kp hat relating to the phase delay used here is given from the equivalent transfer characteristic data 13 described below. In the equivalent transfer characteristic data 13, the phase delay data measured by the preliminary test is stored in advance as the initial value of the convergence stability coefficient G _kp hat regarding the phase delay. This content is
It is sequentially updated by the convergence stability coefficient G _kp hat regarding the phase delay updated by the update algorithm 11.

According to the adaptive control algorithm 1 explained above, the adaptive control test of the periodic signal of the first embodiment is carried out. (Test Facility of Embodiment 1) In this embodiment, the algorithm of the adaptive control method of the periodic signal described above is used for the drastic change of the phase delay of the controlled system by using the experimental electric circuit shown in FIG. On the other hand, it was tested whether the influence of the periodic signal could be suppressed adaptively.

The test equipment of this embodiment is mainly composed of the function generator 20, as shown in FIG.
It is composed of three elements, a phase conversion amplifier 26 and a controller 1. Then, these three elements are connected to each other by a wiring including two 10 kΩ resistors, and a part of the other 10 kΩ
Grounded via an Ω resistor. The function generator 20 is a signal generation source that generates a periodic signal d (n) having a set angular frequency ω ^* . The phase conversion amplifier 26 is an electric circuit that acts as a phase delay element of the controlled system and artificially causes the set phase delay. The controller 1 has an A / D converter on the input side,
It is composed of a personal computer having a D / A converter on the output side and a DSP. This controller 1
Is an algorithm 1 of the adaptive control method for the periodic signal described above.
Is built in as a program and has the ability to process in real time. The above is the outline of the circuit components.

Next, paying attention to the electric signal (voltage) flowing through the circuit, first, the periodic signal d (n) is supplied from the function generator 20 as described above. On the other hand, the controller 1 outputs the adaptive signal y which is the control output.
(N) is supplied according to the above algorithm. The adaptive signal y (n) is converted into a voltage by the D / A converter and output as the adaptive transfer signal z (n) which is the system output. These periodic signal d (n) and adaptive transmission signal z
Both (n) and (n) are combined via an electric resistance of 10 kΩ to form a control target signal s (n). The control target signal s (n) is a signal to be suppressed by the controller 1, and the control target is s (n) = 0, that is, the error signal e (n) = 0.

This control object signal s (n) is given an arbitrary phase delay or phase lead by the phase conversion amplifier 26, A / D converted (sampled) as the error signal e (n), and input to the controller 1. It At the same time, the controller 1 also samples and captures the periodic signal d (n) itself, which is used only for calculating the measured value ω of the primary angular frequency ω ^* . Since this measurement is performed with extremely high precision, it can be safely set as ω = ω ^{* for} engineering purposes.

Now, the signal circuit of the electric circuit configured as described above is represented by the system block diagram shown in FIG. The system shown in the figure differs from the system shown in FIG. 1 in the configuration of the physical system 2 to be controlled. That is, FIG.
In this system, the signal output from the signal generation source 20 and the adaptive signal y (n) are respectively transferred to the signal transfer characteristic 21 (G '
_{g, p} ) and the transfer characteristic G of the controlled system 23
After going through _{g and p} , it becomes a periodic signal d (n) and an adaptive transfer signal z (n). After that, they are combined and the error signal e
(N) = d (n) + z (n) is formed and read into the adaptive control algorithm 1. On the other hand, in the system of FIG.
The periodic signal d (n) from the signal generator 20 which is a function generator and the adaptive signal y (n) output from the adaptive control algorithm 1 are suddenly combined. Combined combined signal s (n) = d (n) + y (n)
Is read into the adaptive control algorithm 1 as an error signal e (n) after passing through the transfer characteristic G _{g, p} of the phase conversion amplifier.

Therefore, there is a slight difference in the definition of the periodic signal d (n) between FIG. 1 and FIG. Further, in FIG. 4, there is nothing corresponding to the adaptive transfer signal z (n) of FIG. 1, and instead, a synthetic signal s (n) is newly introduced. However, in FIG. 1, the signal transfer characteristic 21 (G ′ _{g, p} )
Is equivalent to the transfer characteristic G _{g, p} of the controlled system 23, the combined error signal e (n) is completely equivalent to the error signal e (n) of FIG. Therefore, the role of the phase conversion amplifier 26 in the circuit diagram of FIG. 3 is G _{g, p} in FIG. 4 and G ′ _{g, p} in FIG.
= G _{g, p} is the role of both transfer characteristics.

That is, the phase transfer amplifier 26 incorporated in the circuit of FIG. 3 causes the signal transfer characteristic 21 (G ′ of FIG. 1).
_{g, p} ) and the transfer characteristic G _{g, p of the} controlled system 23 are equivalently set, and the phase delay for each of these angular frequencies can be set arbitrarily. As described above, the test circuit of this embodiment shown in FIG. 3 has a system configuration for carrying out the above-described control method of the present invention.

(Test Results of Embodiment 1) Prior to the test of this embodiment, the initial value of the phase delay G _{k P} hat set in the equivalent transfer characteristic 13 is the control value measured in advance as described above. It is set based on the transfer characteristics of the target physical system 2. The phase delay depends on the A / D converter and D /
It was caused by the characteristics of the A converter, and was a small value that had almost no effect when viewed from the frequency of the periodic signal d (n) to be controlled.

From the function generator 20 in the circuit diagram of FIG. 3, only the fundamental wave, that is, only the first-order sine wave was oscillated in two cases of 30 Hz and 90 Hz. (Thus, the order L of the periodic signal d (n) and the order M of the adaptive signal y (n) are both 1.) In each case, the gain is constant in the phase conversion amplifier 26, and the phase The delay is 90 degrees, 180
The test was conducted by setting four settings of 120 degrees and 270 degrees.

At this time, the step size parameters set in the above equation 26 in the updating algorithm 11 of the adaptive coefficient vector W (n) are positive numerical values, and μ _a = 0.05 μ _p = 10 μ _Gp = 0.05 and the sampling frequency of the controller 1 is 250
It is set to 0 Hz (update cycle T = 1/2500 [s]).

As a result, FIG. 5 (30 Hz) and FIG.
The time response shown at (90 Hz) was obtained. In the figure, the error level refers to the amplitude (voltage) of the error signal e (n), and the output level refers to the amplitude (voltage) of the adaptive signal y (n). In both figures, the horizontal axis indicates the elapsed time up to 3.2 seconds, and the control start time was slightly after 0 seconds, so that the convergence is considerably fast.

Therefore, the test result has two frequencies x
Error signal e for all 8 cases with 4 phase delays
The response of (n) converges, and the time required for convergence is 0.
It was about 5 seconds to 3 seconds. On the other hand, since the output level of the adaptive signal y (n) is adapted to have the same amplitude and opposite phase in synchronization with the periodic signal d (n), naturally the sinusoidal vibration of the same level as the periodic signal d (n) is generated. Stable in.

From this test result, according to the adaptive control method of the periodic signal of this embodiment, even if the variation of the phase delay of the transfer characteristic G _{g, p} of the controlled system reaches 270 degrees, the periodic signal It became clear that the effects could be offset and suppressed. By the way, considering the definition of the periodic signal, a phase delay of 360 degrees is equivalent to zero phase delay. Therefore,
180 degrees (= -180 degrees) and 270 degrees (= -90)
The fact that it could be adapted to the phase delay of (degree) means that the adaptive control succeeded in almost the entire area up to 360 degrees. Therefore, this is a test result suggesting that the adaptive control method of the present embodiment has the ability to adapt to a possibly large phase delay.

Next, the phase delay in the phase conversion amplifier 26 is set to 90 degrees, all other conditions are the same as in the above-mentioned test, and the frequency spectrum concerning the error signal e (n) and the adaptive signal y (n) is set. Analysis was carried out. The data used for the analysis are the data after settling in the steady state, and the data of the transient response are not used, whether the control system (adaptive control algorithm 1) is started or not.

First, the frequency of the periodic signal d (n) is set to 60.
FIG. 7A shows the power spectrum when the control system 1 is not activated and the power spectrum when the control system 1 is activated is shown in FIG. 7B. In FIG. 7 (a),
The error signal e (n) has a peak at 60 Hz,
In FIG. 7B, the peak of the error signal e (n) disappears,
Instead, the output of the adaptive signal y (n) peaks at 60 Hz. In FIG. 7A and FIG. 7B, the error signal e
It can be seen that there is a difference of about 40 dB in the 60 Hz component of (n), and the influence of the error signal e (n) can be suppressed to two orders of magnitude, that is, about 1/100, by the adaptive control method of the present embodiment. .

Next, the frequency of the periodic signal d (n) is changed to 120 Hz, which is doubled.
(B) was obtained. In FIG. 8 (a) where control is not performed, the error signal e (n) has a peak at 120 Hz, but in FIG. 8 (b) where control is applied, the peak of the error signal e (n) disappears, Instead, the output of the adaptive signal y (n) has a peak. Figure 8 (a) without control and Figure 8 with control
With (b), the 120 Hz component of the error signal e (n) is
There is a difference of about 40 dB, and the periodic signal d (n) is obtained by the adaptive control method of the present embodiment as in the case of 60 Hz described above.
It can be seen that the effect of was successfully suppressed.

From the above test results, according to the adaptive control method of the periodic signal of the present embodiment, even for a large change in the phase delay of the controlled system 2 which exceeds 90 degrees and reaches 270 degrees than expected. It was confirmed that the effects of periodic signals can be suppressed by adapting well without divergence. Further, it has been revealed that the suppression capability exerts 40 dB, and the influence of the periodic signal can be suppressed until it is lost by the minute noise components of other frequencies.

[Example 2: Test and evaluation with actual vehicle equipment] As Example 2 of the present invention, an actual vehicle test for suppressing vibration due to the engine on the vehicle floor under the driver's seat of an automobile was carried out. The purpose of this embodiment is to apply the adaptive control algorithm, which has been applied to the ideal periodic signal of only the fundamental vibration in the first embodiment and has achieved good results, to the actual controlled object and confirm the effectiveness. Is. That is, it is effective for periodic signals with almost infinite orders as in many physical systems in reality, and for transient periodic signals in which the primary angular frequency transits further. Two kinds of tests were conducted for the purpose of confirming that two points are also effective.

The present embodiment will be described below with reference to FIGS. 9 to 13. (Control Method of Second Embodiment) As shown in FIG. 9, the overall system image of the adaptive control method of the periodic signal of the present embodiment is similar to that of the first embodiment from the adaptive control algorithm 1 and the physical system 2 to be controlled. Become.

First, the adaptive control algorithm 1 is the same as that of the first embodiment. This is because the purpose of the present embodiment is that the adaptive control algorithm 1 capable of adapting to the large fluctuation of the phase delay of the controlled system in the first embodiment has high-order (order L≈∞) harmonics and the primary angular frequency ω. ^This is because it is to see if the vibration due to the actual engine (in-line 4-cylinder) in which ^* changes can be well suppressed. Here, the order M of the adaptive signal y (n) is 1 as in the first embodiment, because only the secondary component, which is the main vibration component of the engine, is the periodic signal to be controlled. .

Next, the physical system 2 to be controlled is the first embodiment shown in FIG. 1 in the outline of the system block diagram.
The same as the ones, but with some differences: First, the signal generation source 20 is an engine, and its vibration has high-order harmonics, and the primary angular frequency ω ^* fluctuates as the rotational speed changes. Therefore, the periodic signal d (n) that has passed through the signal transfer characteristic 21 formed by the engine mount, the vehicle body, and the like has the transition of the primary angular frequency ω ^* , the amplitude a _k ^* of the fundamental wave and the harmonics, and the phase φ _k ^*. It is necessary to consider it as something to do.

Second, the signal transfer characteristic 21 ( _{G'g, p} )
And the transfer characteristic 23 (G _{g, p} ) of the controlled system are
Not the same. That is, the signal transfer characteristic 21 is a transfer characteristic in which the vibration generated from the signal generation source 20 which is the engine reaches the observation point 24 via the engine mount, the vehicle body, and the like. On the other hand, the transfer characteristic 23 of the controlled system
Is an adaptive signal y (n) input as an electric signal from the adaptive signal generation algorithm 12 drives an actuator incorporated in the engine mount to excite the engine and the vehicle body, and the vibration propagates through the vehicle body and is observed. This is the transfer characteristic up to the point 24.

Thirdly, the measurement of the primary angular frequency ω ^* of the periodic signal d (n) is not performed by observing the periodic signal d (n) as in the first embodiment. Is performed by observing the pulse period of the pulse input from the rotation sensor provided in the engine. Fourth
In addition, since the pickup sensor is provided at the observation point 24, the error signal e which is the output passing to the right in FIG.
Although (n) is mechanical vibration, the error signal e (n) input to the adaptive control algorithm 1 is an electric signal.

Although there are differences from the first embodiment as described above, there is no great difference in the basic configuration of the physical system 2 to be controlled as a system (mathematical model). However, as described above, the periodic signal d (n) has higher-order (order L≈∞) harmonics, and its primary angular frequency ω ^* fluctuates. Is different from that of the first embodiment. Further, the signal transfer characteristics 21 and 23 are not necessarily constant and may change with time. This is also different from the physical system 2 to be controlled in the first embodiment.

(Test Equipment of Embodiment 2) The test equipment used for the actual vehicle test of the adaptive control method of the periodic signal according to the present embodiment is constructed as shown in FIG. That is, this test facility is composed of an automobile (or an in-vehicle system), which is the physical system 2 to be controlled, and the adaptive control device 10 installed on the ground.

First, the physical system 2 to be controlled comprises a normal front engine type automobile and the sensors 24 and 25 / actuator 27 mounted on the automobile. The engine 20, which is a signal generation source, is a gasoline engine with four in-line cylinders. Here, in a four-cylinder engine, since the second-order harmonic vibration is usually the main component of the vibration, the order M of the adaptive control algorithm 1 is set to 1 accordingly.

The camshaft of the engine 20 is equipped with a frequency type rotation sensor 25. The engine speed detected by the rotation sensor 25 (the camshaft speed of 2
Is a real time measurement of the primary angular frequency ω ^* of the periodic signal d (n) to be controlled (measurement value is ω and
Equivalent to ^* ). The engine 20 is supported in the engine compartment of an automobile by an engine mount. Here, of the three engine mounts in total, the engine mount 27 with a built-in actuator occupied one. The engine mount 27 is driven by a built-in actuator by drive power corresponding to the adaptive signal y (n) supplied from the adaptive control device 10, and varies the distance between the engine 20 and the mount support portion of the vehicle,
Vertical vibration is applied to the physical system 2.

A pickup sensor 24 for detecting the vertical acceleration is provided on the floor portion under the driver's seat in the cabin of the automobile. Pickup sensor 24
Is a crystal type piezoelectric effect type accelerometer, from which a voltage as a measurement output can be obtained. The error signal e (n) is picked up here and transmitted to the adaptive controller 10 as an analog electrical signal.

Next, the adaptive control device 10 is configured with the controller 1 incorporating the above-mentioned adaptive control algorithm as the core. As the controller 1, the one used in Example 1 was used as it was. The detector 14 for the angular frequency ω ^* of the periodic signal d (n) to be controlled is a digital signal processor (DSP), and outputs the signal obtained from the output signal from the rotation speed sensor 25 described above. Based on this, the measured value ω of the angular frequency is instantly calculated and supplied to the controller 1 in real time.

The detector 15 for the error signal e (n) is an A / D converter that samples the electric signal output from the pickup sensor 24, digitizes it, and supplies it to the controller 1 in real time. The power amplifier 16 amplifies an electric signal obtained by analogizing the digital adaptive signal y (n) output from the controller 1 by a D / A converter (not shown),
It is supplied to the actuator 27 as driving power which is the adaptive signal y (n).

By the way, the adaptive control system configured as described above cancels the influence of the vibration of the engine 20 and suppresses the vibration at the observation point (pickup sensor 24).
That is, the output of the rotation sensor 25 described above is A / D converted and input to the adaptive control algorithm 1 as the measured value ω of the primary angular frequency of the engine 20. Similarly, the control target vibration at the observation point detected by the pickup sensor 24 is A / D converted and then input to the adaptive control algorithm 1 as an error signal e (n).

Thus, the primary angular frequency ω and the error signal e
The adaptive control algorithm 1 to which (n) is input is executed as a program on a computer, and the adaptive signal y
(N) is output. The adaptive control algorithm 1 gradually adapts as described above and outputs an appropriate adaptive signal y (n). (Test Result of Second Embodiment) In this embodiment, the controlled system G is identified by the vibration test before the control test by the adaptive control algorithm 1. That is, D
Using an SP (digital signal processor), a vibration test was carried out by an automatic frequency sweep (1 Hz step). Then, using the data obtained as a result, the equivalent transfer characteristic data 13 of the controlled system G23 shown in FIG. 9 was obtained again. In this vibration test, both the estimated values G _kg hat and G _kp hat of the gain and the phase delay at each angular frequency, which are the equivalent transfer characteristic data 13, were measured.

As a result, as shown in FIG. 11, both show fairly complicated characteristics. For example, the frequency is 220
At or above Hz, the phase turbulence is significantly disturbed. It is unclear whether this disturbance is due to actual characteristics or due to observation noise, but since there is almost no gain at that frequency, the actual effect can be ignored. Here, since the display of the phase delay is performed between -180 ° and + 180 °, the portion where the data jumps from -180 ° to + 180 ° continuously progresses the phase delay as it is. Want to be regarded as being.

As described above, the gain and the phase delay of the controlled system 23 were measured prior to the test, but the data supplied to the adaptive control algorithm 1 is only the estimated value G _kp hat of the phase delay. By the way, in the adaptive control test of the present embodiment, two types of tests were carried out: a test in the idling state (quasi-steady state) of the engine and a test for observing the transient response during acceleration of the engine. The update cycle (sampling cycle) T and each step size parameter set at that time are the same as those in the first embodiment.

First, in the idling test, the vibration of the vehicle floor due to the engine 20 during idling operation in which the temperature is sufficiently warmed up and the temperature is stable and becomes a quasi-steady state is compared between the case where the adaptive control of the present invention is provided and the case where it is not provided. It measured about two cases. In the measurement, the output of the pickup sensor 24 described above was continuously recorded as numerical data for 20 seconds. The record is processed offline after the measurement is completed,
The power spectrum of acceleration was determined. The sampling cycle was 1/2500 seconds, which is the same as the update cycle T of the adaptive control algorithm 1.

As a result, the power spectrum shown in FIG. 12 was obtained with the solid line with control and the broken line without control. Without control, the fundamental frequency of vibration during idling operation is 23.50 Hz (engine speed 705r
A peak protruding by about 20 to 50 dB from the frequency part before and after pm) is formed. On the other hand, when the control is applied, the above-mentioned peak disappears, and rather the vibration level is lower than that of the front and rear frequencies. Therefore, it has been proved that the adaptive control method of the periodic signal according to the present embodiment can effectively remove the influence of the fundamental vibration even in a real system having higher harmonics.

Next, in the acceleration test, the engine speed was set to 705 rpm for idling (basic frequency 23.5H).
From z) to 6000 rpm (200 Hz) 1 / s
The vibration at the observation point 24 was measured while accelerating at a rate of 00 rpm for about 60 seconds. The measuring means and the sampling cycle at that time are the same as those in the idling test. Also in this test, the case where the control of the present invention was carried out and the case where the control was not carried out were carried out, and the vibration of the acceleration recorded as numerical data was processed offline after the measurement was completed.

As a result, the power spectrum shown in FIG. 13 is obtained, which is several dB to 20d in almost all the rotation speed regions.
A damping effect of about B was observed. In particular, the effect is enhanced in the range of low to medium speeds that are commonly used in ordinary automobiles. Therefore, it was confirmed that the adaptive control method of the periodic signal according to the present example exerts an effective damping effect even on an actual system in which the primary angular frequency ω ^* changes.

As described above, in the two types of actual vehicle tests of this embodiment, according to the adaptive control method of the periodic signal of the present invention, the periodic signal d (n) has higher harmonics and It was confirmed that the vibration of the actual system in which the angular frequency ω ^* fluctuates can also be effectively suppressed. The adaptive signal y
Convergence stability and time required for convergence can be adjusted by appropriately adjusting the order M of (n), the sampling period T, and each step size parameter.

[Embodiment 3: Test and Evaluation in Electric Circuit] As Embodiment 3 of the present invention, the adaptive control method of the periodic signal of the present invention is applied to the same electric circuit as in Embodiment 1, and the phase of the controlled object is applied. It was tested and evaluated whether it could cope with a large variation in delay.
Hereinafter, this embodiment will be described with reference to FIGS. 3 and 4 and FIGS. 14 to 17 again.

(Control Method of Third Embodiment) The adaptive control algorithm of the present embodiment is exactly the same as the adaptive control algorithm 1 of the first embodiment described with reference to FIG. The only difference is that the equivalent transfer characteristic data 13 is not prepared in advance and is set appropriately. That is, the phase delay was set to zero over the entire frequency range.

(Test Equipment of Example 3) The test equipment is the same as the test equipment of Example 1 shown in FIG. (Test Results of Example 3) Similar to Example 1, from the function generator 20 shown in FIG. 3, only the fundamental wave was oscillated in two cases of 30 Hz and 90 Hz.
Then, in each case, the phase conversion amplifier 2
6, the phase delay was set in four ways and the test was conducted. The update cycle T and the step size parameter at that time were also the same as in the first embodiment.

As a result, as in the first embodiment, as shown in FIG.
The time response shown in FIG. 15 (90 Hz) was obtained. According to this, the response of the error signal e (n) converges in all 8 cases with 2 kinds of frequency × 4 kinds of phase delay, and the time required for the convergence is about 0.5 seconds to 1.5 seconds by visual observation. It was extremely quick. From this test result, according to the adaptive control method of the periodic signal of the present embodiment, the influence of the periodic signal is canceled even if the phase delay difference reaches about 300 degrees among the expected transfer characteristics of the controlled system. It was demonstrated that it can be suppressed.

Next, as in the first embodiment, the phase delay in the phase conversion amplifier 26 is set to 90 degrees, all other conditions are the same as in the above-mentioned test, and the error signal e (n) and the adaptive signal y are set.
Frequency spectrum analysis for (n) was performed. First,
FIG. 16 shows power spectra when the frequency of the periodic signal d (n) is set to 30 Hz and the control system is not activated and is activated. Next, the periodic signal d
The frequency of (n) was changed to 90 Hz, which was tripled, and FIG. 17 was similarly obtained. From FIGS. 16 and 17, the same tendency as in Example 1 can be read, and there is a difference of about 40 dB in the peak of the error signal e (n) with and without control. Therefore, similarly to the first embodiment, it is understood that the influence of the periodic signal d (n) can be well suppressed by the adaptive control method of the present embodiment.

From the above test results, according to the adaptive control method of the periodic signal of the present embodiment, it is possible to cope with a large change in the phase delay of the controlled system by about 300 degrees without predicting the transfer characteristic of the controlled system. However, it was possible to demonstrate that the effect of periodic signals can be suppressed by adapting well in a short time. [Example 4: Test and evaluation in actual vehicle equipment] As Example 4 of the present invention, as in Example 2, an actual vehicle test for suppressing vibration due to the engine on the vehicle floor under the driver's seat of the vehicle was performed.

The purpose of this embodiment is to prepare in advance the equivalent transfer characteristic data 13 of the physical system 2 to be controlled by the adaptive control algorithm 1 of the present invention, which has been successfully tested in the actual vehicle test of the second embodiment (FIG. 9). Without thinking, it was to judge whether it can be applied to the actual vehicle. As a result, in this embodiment, the adaptive control system is started with appropriate (for example, all zero) equivalent transfer characteristic data as an initial value, and a periodic signal with almost infinite order and a transition of the primary angular frequency is generated. It can be demonstrated that the effect of can be suppressed well.

Hereinafter, this embodiment will be described with reference to FIGS. 9 and 1 again.
0 and FIGS. 18 and 19 will be described. (Control Method of Fourth Embodiment) The adaptive control method of the periodic signal of the present embodiment is the same as the adaptive control method of the second embodiment derived with reference to FIG. The only difference is that the equivalent control characteristic data 13 of the controlled system has no actual measurement value set in advance, and the adaptive control algorithm 1 is started in the state where the phase delay is initially set to zero over the entire frequency range. Is to be done.

That is, in the updating algorithm 11 for the adaptive coefficient vector W (n), zero is substituted into the convergence stability coefficient G _kp hat regarding the phase delay each time the primary angular frequency ω transits to a new region, and It is required to converge from a state far from the transfer characteristic of the system. However, when entering the region of the same primary angular frequency ω for the second time or later, the value of the convergence stability coefficient G _kp hat updated last time is given to the update algorithm 11.

Therefore, when the control method of the present embodiment can exert good performance, even if the transfer characteristic of the controlled system cannot be predicted at all, as long as the order of the component of the periodic signal to be suppressed is known, the present invention can be realized. The adaptive control of the periodic signal of is applicable. In this way, the significance of the actual vehicle test of this embodiment is great. (Test Facility of Example 4) The test facility of this example is shown in FIG.
This is the same as the test equipment of Example 2 described with reference to FIG.

(Test Results of Example 4) In the adaptive control test of this example, two types of tests, that is, an idling test and an acceleration test, were carried out as in the case of Example 2. The parameters and test conditions at that time are exactly the same as in Example 2. First, in the idling test, the power spectrum shown in FIG. 18 was obtained as in Example 2. If there is no control (broken line), 20-40 dB at the fundamental frequency of 23.50 Hz
A more prominent peak is formed. On the other hand, when the control is applied (solid line), the peak disappears. Therefore, according to the adaptive control method of the periodic signal of the present embodiment,
It has been proved that the influence of the fundamental vibration can be effectively removed even in the real system having high-order harmonics in consideration of the transfer characteristics.

Next, an acceleration test was also conducted under the same test conditions as in Example 2. As a result, the power spectrum shown in FIG. 19 is obtained, which is in the range of several dB in almost all rotation speed regions.
A damping effect of about 20 dB was observed. Therefore, according to the adaptive control method of the periodic signal of the present embodiment, even in the actual system in which the primary angular frequency ω ^* transits, the effective damping effect for the prediction regarding the transfer characteristic is exerted. It was proved. Here, as described above, the damping effect of about 20 to 40 dB was exhibited for the constant primary angular frequency ω ^* , so that the primary angular frequency ω ^* does not change transiently. It is considered that the same damping effect can be obtained if the rotation speed is settled at an arbitrary speed within the range of the acceleration test.

In the second embodiment, the measurement data of the equivalent transfer characteristic of the controlled system is preset and then updated by the convergence stability coefficient. In this embodiment, the measurement data is updated by the convergence stability coefficient from the initial value of zero. Therefore,
In Example 2 in which a good initial value is given, the control performance is slightly better in the first sweep. As mentioned above,
In the actual vehicle test of this embodiment, according to the adaptive control method of the periodic signal of the present invention, the vibration of the actual system having high-order harmonics and varying the primary angular frequency is also related to the transfer characteristic of the controlled system. It has been demonstrated that it can be effectively suppressed without the need for any preliminary information.

(Modification of Fourth Embodiment) In this embodiment, it is not always necessary to preset the equivalent transfer characteristic data 13 of the controlled system in the adaptive control algorithm 1 shown in FIG. I was killed. Therefore, the convergence stability coefficient G _k hat, which is a component of the adaptive coefficient vector W (n) as well as the amplitude a _k and the phase φ _{k of} the adaptive signal y (n), is stored in the update algorithm 11, and the primary angle Equivalent transfer characteristic data 1 even if the frequency ω changes
It is also possible to construct an adaptive control algorithm that does not need to be supplied from the 3rd party.

That is, as shown in FIG. 20, the equivalent transfer characteristic data (13 in FIG. 9) of the controlled system is omitted, and the update algorithm 11 for the adaptive coefficient vector W (n) is set.
It is possible to configure the adaptive control algorithm 1 only with the adaptive signal generation algorithm 12. Since this adaptive control algorithm 1 does not store the equivalent transfer characteristics of the controlled system 23 as a data table, the control results will not improve no matter how many times the same angular frequency is experienced. However, the memory capacity required for storing the equivalent transfer characteristic data becomes unnecessary. Also, a step of reading new data from the memory each time the primary angular frequency ω changes,
The step of updating the equivalent transfer characteristic data from the updated convergence stability coefficient G _k hat every time the adaptive coefficient vector W (n) is updated is omitted.

As a result, the adaptive control algorithm of this structure has a very simple logic structure, and the number of steps of the program to be executed and the memory capacity to be secured are small. Therefore, according to the adaptive control algorithm of the present configuration, it is possible to provide an inexpensive, lightweight and small adaptive control system for a periodic signal while having the ability to adapt to the frequency change of the periodic signal and the change of the transfer characteristic of the controlled system. It will be possible.

[Possibility of Incorporation of this Control System into Vehicle] The adaptive control method of the present invention, which has been illustrated as the above-described second and fourth embodiments and its modification, is installed in an actual vehicle as a damping system incorporating the same. There is a possibility. That is, first, an actual vehicle test is repeated using the above-mentioned equipment, and an adaptive control algorithm having sufficient performance and having the lowest cost of the control device is found. And, only the minimum required functions of ground equipment are compactly packed into one semiconductor circuit or one board, and combined with power amplifier, A / D converter, D / A converter, etc. When the two units are mounted together on a vehicle, they can be commercialized as an on-vehicle system that implements the adaptive control method of the periodic signal of the present invention.

At this time, considering the electromagnetic noise environment, A /
The D converter may be provided near the sensor. Further, the power amplifier that generates a large amount of heat may be provided outside the above-mentioned control unit to achieve good cooling. In the unlikely event that the control system does not function, it is desirable to provide a vibration damping device that is not comfortable but can be operated for the time being.

Finally, the adaptive control method of the periodic signal according to the present invention not only suppresses the vibration and noise of the automobile but also
Helicopters, propeller planes, ships (including submarines), etc. where vibration and rotation noise due to rotors in the cabin become a problem,
It can be applied to many vehicles. Similarly, it can be applied to various machine tools, various electric circuits, and vibration control and noise reduction in buildings.

[0153]

As described in detail above, according to the adaptive control method of the periodic signal of the present invention, it is possible to adapt to a large change in the transfer characteristic (gain and phase delay at each frequency) of the system to be controlled. By following this, it is possible to suppress or remove the influence of the periodic signal at the observation point. In addition, according to the adaptive control method of the periodic signal of the present invention, the influence of the periodic signal can be sufficiently suppressed by adapting to the large change of the fundamental frequency of the periodic signal whose effect should be removed. it can.

That is, according to the present invention, the transfer characteristic data equivalent to the system to be controlled is used as an initial value, and the value is updated so that the entire system is always stable. We provided an adaptive control algorithm that can adapt to large changes in phase lag. Not only that, there is no need to measure the transfer characteristics of the system to be controlled in advance, and an adaptive control algorithm that automatically follows changes in the transfer characteristics could be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing a general form of an adaptive control method of the present invention.

FIG. 2 is a phase diagram of ψ showing a convergence range of the adaptive control method of the present invention.

FIG. 3 is a test circuit diagram of the adaptive control system according to the first embodiment.

FIG. 4 is a block diagram equivalent to an experimental circuit of the adaptive control method of the first embodiment.

FIG. 5 is a time response of adaptive control with respect to the periodic signal (30 Hz) of the first embodiment.

FIG. 6 is a time response of adaptive control for the periodic signal (90 Hz) of the first embodiment.

7 is a frequency spectrum for the periodic signal (60 Hz) of Example 1. FIG.

8 is a frequency spectrum for the periodic signal (120 Hz) of Example 1. FIG.

FIG. 9 is a block diagram showing an adaptive control method according to a second embodiment.

FIG. 10 is a schematic diagram showing the configuration of the test equipment of the adaptive control system of the second embodiment.

FIG. 11 is a Bode diagram showing measured transfer characteristics of the controlled system of the second embodiment.

FIG. 12 is a power spectrum of an error signal during idling in the second embodiment.

FIG. 13 is a power spectrum of an error signal during acceleration according to the second embodiment.

FIG. 14 is a time response of adaptive control with respect to a periodic signal (30 Hz) of Example 3.

FIG. 15 is a time response of adaptive control with respect to a periodic signal (90 Hz) of Example 3.

16 is a frequency spectrum for a periodic signal (30 Hz) of Example 3. FIG.

FIG. 17 is a frequency spectrum for the periodic signal (90 Hz) of the third embodiment.

FIG. 18 is a power spectrum of an error signal during idling in the fourth embodiment.

FIG. 19 is a power spectrum of an error signal during acceleration according to the fourth embodiment.

FIG. 20 is a block diagram showing a modified adaptive control method.

FIG. 21 is a block diagram showing an adaptive control method of the related art (FX-LMS).

FIG. 22 is a block diagram showing a prior art (modified FX-LMS) adaptive control method.

FIG. 23 is a block diagram showing a prior art (SFX) adaptive control method.

FIG. 24 is a block diagram showing a prior art (DXHS) adaptive control method.

FIG. 25 is a block diagram showing a test facility for electric circuits of the related art and the prior art.

FIG. 26 is a diagram illustrating a conventional technique (FX-L for a change in phase delay).
MS) time response

FIG. 27: Prior art (DXH for changes in phase delay)
S) time response

[Explanation of symbols]

1: Adaptive control algorithm (controller, computer, control system) 10: Adaptive control device 11: Adaptive coefficient vector W
(N) update algorithm 12: Adaptive signal generation algorithm (y (n) = ...) 13: Equivalent transfer characteristic data (Gk) of controlled system G
Hat) 14: Primary angular frequency ω detection means (signal processing device) 15 of the periodic signal d (n) 15: Error signal e (n) detection means (signal processing device)
16: Power amplifier 2: Physical system to be controlled (electric circuit / automobile) 20: Periodic signal generation source (function generator / engine) 21: Transfer characteristic of periodic signal 22: Periodic signal generation system 23: Control target system (Transfer characteristic is G _{g, p} ) 24: Error signal e (n) observation point (voltage sensor / pickup sensor) 25: Primary angular frequency ω measuring means of periodic signal d (n) (rotation speed sensor) 26: Phase Conversion amplifier 27: Engine mount with built-in actuator d (n): Periodic signal e (n): Error signal y
(N): Adaptive signal z (n): Adaptive transmission signal (synthesizer output / actuator transmission output) n: Time (step) T: Update cycle (sampling cycle) W (n): Adaptive coefficient vector a _k ^* , φ _k ^* : Amplitude / phase of k-th order sine wave of periodic signal d (n) (1 ≦ k ≦ L) a _k , φ _k : Amplitude and phase (1) of k-th order sine wave of adaptive signal y (n) ≦ k ≦ M) L: Maximum order of higher-order vibration of the periodic signal to be controlled (1 ≦
L) M: maximum order of higher-order vibration of adaptive signal (1 ≦ M ≦ L) ω ^* : primary angular frequency of periodic signal d (n) ω: primary angular frequency of adaptive signal y (n) (ω ^* Measured value of ^* is equivalent to ω ^{* in} terms of engineering) G g, p: Transfer characteristic of controlled system G'g, p:
Transfer characteristics of periodic signal G _kg , G _kp : Gain and phase delay at angular frequency kω of controlled system G _k hat: Convergence stability coefficient (at least one of G _kg hat and G _kp hat) G _kg hat, G _kp hat: Convergence stability coefficient for gain and phase delay ψ: Phase adjustment parameter

Continuation of the front page (56) Reference JP-A-5-61485 (JP, A) JP-A-6-282273 (JP, A) JP-A-6-43882 (JP, A) JP-A-7-77988 (JP , A) JP-A-8-44377 (JP, A) International Publication 94/024970 (WO, A1) The Institute of Electronics and Communication Engineers, Digital Signal Processing, Japan, The Institute of Electronics and Communication Engineers, August 10, 1980, 8th edition. , P. 228, 229 (58) Fields surveyed (Int.Cl. ⁷ , DB name) G10K 11/178 F16F 15/02 G05B 13/02

Claims (5)

(57) [Claims] 1. For a periodic signal affecting an observation point, a first-order basic sine wave synchronized with the periodic signal and / or the basic sine wave to the M-th order (2 ≦ M) of the basic sine wave. A method for adaptively controlling a periodic signal, which actively removes an influence of a specific frequency component of the periodic signal on the observation point by adding an adaptive signal composed of a harmonic signal of , An adaptive signal generation algorithm for generating the adaptive signal y (n) based on the primary angular frequency of the periodic signal , and each order of the adaptive signal y (n) (order k
= 1, 2, ..., M) sinusoidal amplitude a _k and phase
φ _k as well as the convergence relates to a phase lag stability coefficient G _{k [p]} Hatton
Of the adaptive coefficient vector W (n) whose component is
Error signal detected at the observation point whose signal effect should be removed
According to Equation 2 including the phase adjustment parameter ψ based on e (n)
Therefore, each component of the adaptive coefficient vector W (n) is adaptively adjusted with respect to the amplitude and phase of the periodic signal and the transfer characteristic of the controlled system by updating each time n. And an adaptive coefficient vector updating algorithm for updating the adaptive signal y with an amplitude a _k and a phase φ _k of each order that is a part of the components of the updated adaptive coefficient vector.
(N) The amplitude a _k and the phase φ _k of the sine wave of each order in (n) are updated. [Equation 1] [Equation 2] 2. The adaptive coefficient vector update algorithm according to claim 2, wherein the phase adjustment parameter is set to ψ = π / 2.
2. The adaptive control method for a periodic signal according to claim 1, wherein the adaptive control method is updated according to the equation (3). [Equation 3] 3. An estimated value of a phase delay corresponding to each angular frequency of a transfer characteristic from the adaptive signal y (n) to the observation point.
And the convergence stability coefficient G _{k [p]} hat corresponding to each order that is a component of the adaptive coefficient vector W (n) has the following equivalent transfer characteristic data _: The adaptive control method for a periodic signal according to claim 1, wherein an initial value is given from the equivalent transfer characteristic data corresponding to the angular frequency of each order. 4. The adaptive control method for a periodic signal according to claim 3, wherein the equivalent transfer characteristic data is preset before starting adaptive control. 5. The periodic signal according to claim 3, wherein the equivalent transfer characteristic data is updated by the convergence stability coefficient G _{k [p]} hat in the updated component of the adaptive coefficient vector W (n). Adaptive control method.