JPH10301579A - Adaptive control system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an adaptive control system capable of improving adaptive control performance and obtaining the stable adaptive control performance for a long time. SOLUTION: This system is constituted as follows: a transfer function H' estimating an operation characteristic of an interference wave motion generation means is optimized successively by the first adaptive control algorithm LMSH based on a control signal y outputted from an adaptive filter and a detection signal e of the first detection means, and the transfer function C' of the adaptive filter is optimized successively by the second adaptive control algorithm so as to follow up a change in a propagation characteristic of a wave motion from a wave motion generation source to a control point based on output information r making the detection signal of the second detection means of this transfer function to be an input and the detection signal e of the first detection means, and the control signal y according to the input of the detection signal of the second signal detection means is outputted to the interference wave motion generation means based on the transfer function C' of this adaptive filter.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an adaptive control system suitable for controlling, for example, noise in a vehicle or vibration of a vibration isolation table.

[0002]

2. Description of the Related Art In recent years, attempts have been made to suppress noise transmitted from an engine in a passenger compartment of a vehicle or the like. For example, in order to suppress noise at a predetermined position in the passenger compartment, a noise control device is provided in the passenger compartment. 2. Description of the Related Art There is known a method of actively suppressing noise by applying a sound wave having an amplitude and a phase corresponding to the noise from a speaker to a predetermined position and causing the sound to interfere with the noise. For example, a method has been proposed in which noise can be continuously suppressed by sound waves from the speaker by adaptively controlling the output of sound waves from the speaker, even if the propagation characteristics of the noise propagation path fluctuate. . Here, FIG. 4 shows a configuration example of a vehicle interior noise suppression system using adaptive control. In FIG. 4, a microphone 11 for detecting noise from the engine and converting the noise into an electric signal is provided in the vicinity of the engine of the vehicle 2, and the noise in the vicinity of the seat is detected and converted to an electric signal in a seat in the vehicle compartment. Microphone 13 is provided, and a speaker 15 capable of outputting a sound wave is provided on a ceiling portion in the vehicle cabin. The adaptive control apparatus 101 receives the output x of the microphone 11 and the output e of the microphone 13 and outputs a sound signal y to the speaker 15 such that the output e of the microphone 13 is minimum and ideally zero. When the sound wave is output from the speaker 15, the noise from the engine and the sound wave from the speaker 15 interfere and cancel each other near the seat, and the noise in the vehicle compartment is suppressed.

FIG. 5 shows an example of a model of a vehicle interior noise suppression system using adaptive control configured as described above. In the model shown in FIG. 5, a detection signal x of a microphone 11 for detecting engine noise is input and a noise signal e at a seat position detected by the microphone 13 is output, and C is a seat of the noise x emitted from the engine. , K is a transfer function indicating the operating characteristics of the speaker 15, K ′ is an estimated value of the transfer function of the speaker 15, H is the transfer function of the adaptive filter, and LMS (Least Mean Meaning). Squar
e) shows a predetermined adaptive control algorithm. Also,
Adaptive filter transfer function H, adaptive control algorithm LM
The adaptive control device 101 realizes S and the estimated value K ′ of the transfer function of the speaker 15.

In such a configuration, the transfer function H of the adaptive filter minimizes the amplitude of the detection signal e of the microphone 13 (ideally, from the output z from the transfer function K ′ and the information of the detection signal e of the microphone 13). The coefficient parameter of the adaptive filter is updated by the adaptive control algorithm LMS so as to output a control signal y to the speaker 15 to be zero. The update of the transfer function H of the adaptive filter is sequentially performed based on, for example, the following equation.

(Equation 1) Here, μ _H is a coefficient for adjusting the convergence characteristic of the transfer function H of the adaptive filter. Note that the above equation is a so-called filter
Called the ed-X algorithm.

The transfer function H of the adaptive filter is determined by the transfer function K ′ of the estimated speaker 15 and the microphone 13
Even if the transfer function C indicating the propagation characteristic of the propagation path fluctuates on the basis of the information of the detection signal e detected by the above, it can follow this. Therefore, the control signal y from the transfer function H of the adaptive filter, which is sequentially updated, is input to the speaker 15 represented by the transfer function K, and as a result, the speaker 15
From the engine, interferes with noise from the engine that has propagated along the propagation path represented by the transfer function C near the seat, cancels each other, and suppresses the noise in the passenger compartment.

[0006]

However, the adaptive control model shown in FIG. 5 has a configuration in which a transfer function K 'for estimating the operating characteristics of the speaker 15 is identified in advance and adaptive control is performed. Further, the actual operating characteristics of the speaker 15 may fluctuate due to aging or the like, and the transfer function K may fluctuate. When the actual operation characteristics of the speaker 15 fluctuate, the adaptive control algorithm LMS uses the transfer function H ′ of the adaptive filter based on the information of the estimated transfer function K ′ different from the actual transfer function and the detection signal e detected by the microphone 13. Therefore, there is a disadvantage that the noise suppression performance of the above system is not stable.

The present invention has been made in view of the above problems, and an object of the present invention is to provide an adaptive control system capable of improving adaptive control performance and obtaining stable adaptive control performance for a long time. Aim.

[0008]

An adaptive control system according to the present invention comprises: first detection means for detecting a wave propagating to a control point; second detection means for detecting a wave emitted from a wave generation source; An interfering wave generating means for generating an interfering wave for canceling the wave transmitted from the wave generating source to the control point through a predetermined path, and inputting a detection signal from the second detecting means to generate the wave. A control signal for causing the interference wave generating means to generate an interference wave that minimizes the amplitude of the wave at the control point in accordance with a change in the propagation characteristic of the wave on the path from the source to the control point; Adaptive control means having an adaptive filter for outputting the adaptive filter, the adaptive control means based on a control signal output from the adaptive filter and a detection signal of the first detection means, A transfer function for estimating an operation characteristic of the interference wave generating means is sequentially optimized by a first adaptive control algorithm, and output information of the transfer function using a detection signal of the second detection means as input and the first detection information A transfer function of the adaptive filter is sequentially optimized by a second adaptive control algorithm so as to follow a change in a propagation characteristic of a wave from the wave generation source to a control point based on the detection signal of the adaptive filter. The control signal corresponding to the input of the detection signal of the second signal detection means is output to the interference wave generation means based on the transfer function of the second signal detection means.

In the present invention, when sequentially specifying the transfer function of the adaptive filter, first, the transfer function for estimating the operation characteristic of the interference wave generating means is sequentially optimized by the first adaptive control algorithm. Next, the transfer function of the adaptive filter is changed based on the output information of the transfer function for estimating the operation characteristic of the interference wave generating means, which is the detection signal of the second detection means, and the detection signal of the first detection means. The second adaptive control algorithm sequentially optimizes the change in the propagation characteristic of the wave from the generation source to the control point. As a result, even if the actual operating characteristics of the interference wave generating means fluctuate due to aging, etc., the transfer function for estimating the operating characteristics of the interference wave generating means necessary for sequentially specifying the transfer function of the adaptive filter is determined by the interference. By following the fluctuation of the operation characteristics of the wave generating means, adaptive control performance can be improved, and stable adaptive control performance can be obtained for a long time.

[0010]

Embodiments of the present invention will be described below with reference to the drawings. First Embodiment FIG. 1 is a configuration diagram when an adaptive control system according to an embodiment of the present invention is applied to suppression of engine noise in a vehicle compartment. In FIG. 1, a microphone 11 for directly detecting noise from the engine 4 and converting the noise into an electric signal is provided near an engine 4 installed in an engine room of the vehicle 2, and a seat near the seat is provided in a seat in the vehicle interior. A microphone 13 that detects noise and converts the noise into an electric signal is provided, and a speaker 15 that can output a sound wave is provided on a ceiling portion of the vehicle interior. Since the microphone 11 is installed in the engine room of the vehicle and the microphone 13 is provided in the vehicle interior, the sound wave output from the speaker 15 captured by the microphone 11 is negligible.

The adaptive control device 21 is installed at a predetermined position of the vehicle 2 and has the above-described microphones 11 and 13.
Are fed back, and a control signal y is output to the speaker 15. The adaptive control device 21 has a microprocessor 23, A / D converters 25 and 27, and a D / A converter 29.

The microprocessor 23 is a circuit that performs various operations and implements an adaptive filter. The A / D converter 27 is a circuit that converts the detection signal x of the microphone 11 into a digital signal and outputs the digital signal to the microprocessor 23. The A / D conversion unit 25 includes the microphone 13
Is a circuit for converting the detection signal e into a digital signal and outputting it to the microprocessor 23. D / A converter 29
Is a circuit that converts a digital control signal output from the microprocessor 23 into an analog signal and outputs the analog signal to the speaker 15.

FIG. 2 is an explanatory diagram showing a model including an internal model of the adaptive control device of the noise suppression system of FIG.
Hereinafter, the noise suppression system according to the present embodiment will be described with reference to FIG. The model shown in FIG. 2 receives a detection signal x of a microphone 11 for detecting engine noise as an input, and outputs a noise signal e at a seat position detected by the microphone 13, and C is provided on a seat for noise from the engine. H is a transfer function indicating the operating characteristics of the speaker 15, H 'is an estimated value of the transfer function of the speaker 15, C' is the transfer function of the adaptive filter, and LMS _H is the first. LMS _C indicates the second adaptive control algorithm. Further, the transfer function C ′ of the adaptive filter, the first and second adaptive control algorithms LM
S _H , LMS _C and the estimated value H ′ of the transfer function of the speaker 15 are realized by the adaptive control device 21.

In FIG. 2, an output y (control signal) of a transfer function C ′ of an adaptive filter to which a detection signal x of a microphone 11 is input is input to a first adaptive control algorithm LMS _H , and the microphone 13 Is input. First adaptive control algorithm LM
S _H updates the estimated value H ′ of the transfer function of the speaker 15 based on the detection signal e of the microphone 13 and the output y of the transfer function C ′. The estimated value H ′ of the transfer function of the speaker 15 can be specified by previously identifying the actual operation characteristics of the speaker 15 by a predetermined identification method.

Specifically, the estimated value H 'of the transfer function of the speaker 15 is updated by the control algorithm shown in the following equation.

(Equation 2) Here, μ _H is a coefficient for adjusting the convergence characteristic of the estimated value H ′ of the transfer function of the speaker 15. According to the above expression, the estimated value H ′ of the transfer function of the speaker 15 is updated so as to converge to the optimum value at that time. Therefore, even if the operating characteristics of the speaker 15 fluctuate due to deterioration over time or the like,
The estimated value H ′ of the transfer function of the loudspeaker 15 follows the fluctuation of the operating characteristics of the loudspeaker 15.

The second adaptive control algorithm LMS _C receives the output r of the updated estimated transfer function H ′ of the speaker 15 and the detection signal x of the microphone 11, and detects the microphone 13. Signal e
Is entered. Second adaptive control algorithm LMS _C
Are the detection signal x of the microphone 11 and the speaker 1
5, the transfer function C ′ of the adaptive filter is updated based on the output r of the estimated value H ′ of the transfer function.

Specifically, the transfer function C 'of the adaptive filter is updated by the control algorithm shown in the following equation.

(Equation 3) Here, μ _C is a coefficient for adjusting the convergence characteristics of the transfer function C ′ of the adaptive filter. According to the above expression, the transfer function C ′ of the adaptive filter is updated so as to converge to the optimum value at that time, and the coefficient parameter of the adaptive filter is updated. Therefore, the transfer function C of the transfer path of the noise from the engine to the microphone 13 provided on the seat is provided.
Is changed due to an environmental change or the like, the transfer function C ′ of the adaptive filter follows this.

From the adaptive filter that has been successively updated, the output y corresponding to the input detection signal x of the microphone 11 is output.
Is output to the speaker 15 as a control signal. The loudspeaker 15 outputs a sound wave whose amplitude and phase are adjusted with respect to the wave generation source so that the amplitude of the detection signal e detected by the microphone 13 is minimized (ideally zero).

As a result, the noise of the engine that reaches the microphone 13 after passing through the transfer function C and the sound wave output from the speaker 15 interfere with each other near the microphone 13 and cancel each other.
That is, noise from the engine near the seat is suppressed.

Further, since the estimated value H 'of the transfer function of the speaker 15 can follow the fluctuation of the operating characteristic of the speaker 15, the estimated value H' of the transfer function of the speaker 15 and the true value H of the operating characteristic of the speaker 15 can be obtained. Is always within a substantially constant range. Therefore, in the noise suppression system according to the present embodiment, even if there is a considerable difference between the initial value of the estimated value H ′ of the transfer function of the speaker 15 and the transfer function H of the speaker 15, The function estimated value H ′ converges on the transfer function H of the speaker 15 while being updated. As a result, the noise suppression performance of the noise suppression system according to the present embodiment is improved.

Further, even if the operating characteristics of the speaker 15 fluctuate due to deterioration over time or the like, the estimated value H 'of the transfer function of the speaker 15 can follow the fluctuation of the operating characteristics of the speaker 15, and The transfer function C ′ of the adaptive filter is sequentially updated on the basis of the output r of the estimated value H ′ and the detection signal e of the microphone 13, so that the transfer function C ′ of the adaptive filter is the transmission of noise from the engine to the microphone 13. It is possible to accurately follow the variation of the transfer function C of the path. As a result, even if the transfer function C of the transfer path of the noise from the engine to the microphone 13 and the transfer function H indicating the operating characteristics of the speaker 15 fluctuate with time, the noise suppression system according to the present embodiment has a long-term stable noise. Can be suppressed.

Second Embodiment Next, FIG. 3 is a configuration diagram in the case where the adaptive control system according to the present embodiment is applied to a vibration suppression system of a vibration isolation table. The anti-vibration table 51 is used, for example, to install equipment such as a stepper used for manufacturing a semiconductor device, which should minimize vibration. The vibration isolation table 51 shown in FIG. 3 basically includes a table 52 on which a target device is installed, legs 54 supporting the table 52, and an installation table 53 connected to the legs 54 and installed on the installation surface G. Is done. An actuator 55 capable of generating vibrations of any frequency, phase, and amplitude is fixed to the table 52, and an acceleration sensor 57 that detects vibrations generated in the table 52 and converts the vibrations into electric signals is provided. . The installation table 53 is provided with an acceleration sensor 56 that detects vibration transmitted to the installation table 53 and converts the vibration into an electric signal. Further, the actuator 55 and the acceleration sensors 56 and 57 are connected to the adaptive control device 21. The adaptive control device 21
The configuration is the same as that described in the first embodiment.

The anti-vibration table 51 usually includes a leg 54 and an installation table 5.
3, a damping mechanism by air, for example, is provided. The damping mechanism attenuates the vibration transmitted from the installation surface G, and the table 52
Passively suppresses vibration transmitted to the vehicle. However,
In the passive vibration suppression method as described above, the table 52
May not be sufficient. For this reason, the vibration isolation table 5
By applying the adaptive control system according to the present invention to 1, the vibration for canceling the vibration transmitted to the table 52 is generated in the actuator 55, thereby actively removing the vibration of the vibration isolation table 51. .

The above-described vibration suppression system is modeled by setting a transfer function indicating a transfer characteristic of a transmission path of a vibration transmitted from the installation surface G to the table 52 through the installation table 53 and the leg 54 as C, and operating the actuator 55 It is assumed that a transfer function indicating a characteristic is H, a detection signal of an acceleration sensor 56 for detecting vibration transmitted to the mounting table 53 is input x, and a detection signal of an acceleration sensor 57 for detecting vibration of the table 52 is output e. By performing such modeling, a model having exactly the same configuration as the model shown in FIG. 2 is obtained.

Therefore, when the adaptive control of the actuator 55 is performed by the adaptive control device 21, the vibration control of the table 52 is performed in the same manner as in the vehicle interior noise suppression system shown in the first embodiment. . As a result, the vibration isolation performance of the table 52 is improved, and the vibration isolation performance of the table 52 is stably obtained for a long time.

[0026]

As described above, according to the adaptive control system of the present invention, the adaptive control performance can be improved, and the adaptive control performance that is stable for a long time can be obtained.

[Brief description of the drawings]

FIG. 1 is a configuration diagram when an adaptive control system according to an embodiment of the present invention is applied to suppression of engine noise in a vehicle cabin.

FIG. 2 is an explanatory diagram showing a model including an internal model of an adaptive control device of the noise suppression system of FIG. 1;

FIG. 3 is a configuration diagram when the adaptive control system according to the embodiment of the present invention is applied to vibration suppression of a vibration isolation table.

FIG. 4 is an explanatory diagram showing one configuration example of a vehicle interior noise suppression system using adaptive control.

FIG. 5 is an explanatory diagram showing an example of a model of a vehicle interior noise suppression system using adaptive control.

[Explanation of symbols]

 2 vehicle 4 engine 11 13 microphone 15 speaker 21 adaptive control device 23 microprocessor 25, 27 A / D converter 29 D / A converter

Claims (1)

[Claims] A first detecting means for detecting a wave propagating to a control point; a second detecting means for detecting a wave generated from a wave generating source; An interference wave generating means for generating an interference wave for canceling the wave propagated to the control point; and a propagation of a wave on a path from the wave generation source to the control point by input of a detection signal of the second detection means. Adaptive control means having an adaptive filter for outputting, to the interference wave generation means, a control signal for causing the interference wave generation means to generate an interference wave that minimizes the amplitude of the wave at the control point according to a change in the characteristic. The adaptive control means, based on a control signal output from the adaptive filter and a detection signal of the first detection means,
A transfer function for estimating an operation characteristic of the interference wave generating means is successively optimized by a first adaptive control algorithm, and output information of the transfer function using a detection signal of the second detection means as input and the first detection information A transfer function of the adaptive filter is sequentially optimized by a second adaptive control algorithm so as to follow a change in a propagation characteristic of a wave from the wave generation source to a control point based on the detection signal of the adaptive filter. An adaptive control system for outputting a control signal corresponding to the input of the detection signal of the second signal detecting means to the interference wave generating means based on the transfer function of the second signal detecting means.