JP2009061912A - Noise reduction device and its method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a noise reduction device etc., hardly increasing noise even if test vibration is generated. <P>SOLUTION: The device comprises a plurality of acceleration sensors for detecting the vibration of a vehicle structure, a plurality of piezo actuators for applying wave motion to the vehicle structure, an abnormality detection part 43 for detecting abnormalities of the acceleration sensors based on the vibration of the vehicle structure detected by the acceleration sensor when the test vibration is generated by the piezo actuator, and a noise control part 42 operating the piezo actuators so as to reduce noise in a predetermined space according to the vibration detected by a plurality of the acceleration sensors when abnormalities of the piezo actuators are not detected by the abnormality detection part 43. The abnormality detection part 43 generates the test vibration by at least one acceleration sensor and reduces the noise generated in the predetermined space caused by the test vibration by a piezo actuator other than the piezo actuator which has generated the test vibration. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a noise reduction apparatus and method for reducing noise in a vehicle interior by generating a wave such as vibration or sound.

  2. Description of the Related Art Conventionally, a noise control device or the like that reduces noise by measuring noise in a passenger compartment or the like generated as a vehicle travels and generating sound waves that cancel the noise has been proposed.

  This noise control device is provided with a plurality of vibration detection means (sensors) for detecting vibrations of the vehicle body, and operates actuators such as speakers and vibrators installed in the vehicle based on the detected vibrations of the vehicle body. Reduce noise.

In such a noise control device, it becomes difficult to effectively reduce noise when an abnormality occurs in the vibration detection sensor. For this reason, conventionally, in the following Patent Document 1 or the like, a microphone for effect confirmation for measuring the noise reduction effect is installed in a place where noise control is performed, and it is determined whether or not effective noise control is actually performed. It is described. And in this patent document 1, when abnormality of a vibration detection sensor generate | occur | produces as a result of this determination and the noise reduction effect is no longer sufficient, the control content for reducing noise is changed, and it is more effective. A technique for noise control is described.
Japanese Patent Laid-Open No. 8-2922771

  However, the above-described conventional noise control device cannot generate an appropriate wave for reducing noise when an abnormality occurs in an actuator that generates vibration in order to reduce noise. It is difficult to do it automatically.

  For this reason, conventionally, vibration for testing the operation of the actuator is generated to detect abnormality of the actuator. However, with this technology, there is a possibility that new noise is generated due to vibration for testing.

  Therefore, the present invention has been proposed in view of the above-described circumstances, and an object thereof is to provide a noise reduction apparatus and method that does not increase noise even when test vibration is generated.

  In order to solve the above-described problem, a noise reduction device according to the present invention is a vehicle structure detected by a vibration detection unit when a vibration for test is generated by a wave applying unit that applies a wave to the vehicle structure. An abnormality detecting means for detecting an abnormality of the wave applying means based on the vibration generates a test vibration by at least one wave applying means, and the wave applying means other than the wave applying means that has generated the test vibration Noise generated in a predetermined space due to test vibration is reduced.

  According to the noise reduction device of the present invention, the test vibration is generated by at least one wave application unit, and the test vibration is caused by the wave application unit other than the wave application unit that generated the test vibration. Since the noise generated in the predetermined space is reduced, it is possible to avoid increasing the noise in the predetermined space in order to detect the abnormality of the wave applying means.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[First Embodiment]
For example, as illustrated in FIG. 1, the present invention is applied to a noise reduction device that reduces noise in a control space P that surrounds a passenger's head in a seat in a vehicle interior surrounded by a vehicle structure. In order to detect an abnormality of a piezo actuator (wave applying means) that performs an operation to reduce noise, the noise reduction device generates a test vibration and performs an abnormality determination process on the piezo actuator, and also performs the test vibration. Simultaneously with the generation of the noise, the noise generated by the test vibration is canceled.

  First, a general technique for reducing noise in the control space P in the noise reduction device according to the first embodiment to which the present invention is applied will be described, and then a characteristic configuration of the noise reduction device to which the present invention is applied will be described. explain.

  The noise reduced by the noise reduction device is vehicle interior noise that enters the vehicle interior from outside the vehicle. Typical causes of vehicle interior noise are engine noise due to engine vibration, road noise noise caused by road surface unevenness entering the floor panel from the tire during travel, and air noise during travel. There is wind noise generated by the air current.

  The noise reduction device according to the present invention is mainly intended to reduce road noise. 1 to 4 show main propagation paths of vehicle body vibration and road noise caused by road surface unevenness. As shown in FIG. 2, the vibration that is the main component of road noise that has entered the vehicle body from the tire 2 first enters a highly rigid beam-like member called the member 103 from the attachment point of the axle 101 and the suspension 102. Thereafter, vibration propagates to a plate-like member called floor panel 1 surrounded by the member 103 and having relatively low rigidity, and the floor panel 1 vibrates. Furthermore, air vibrations in the passenger compartment are caused by the membrane vibration of the floor panel 1 and a resonance phenomenon occurs in the passenger compartment, so that road noise noise is heard in a predetermined space (control space) P in the passenger compartment. Although noise is generated when the roof panel and the window glass vibrate in addition to the floor panel 1, most of the road noise mainly entering from the attachment portion of the suspension 102 is caused by the vibration of the floor panel 1. I know that.

  Therefore, as shown in FIGS. 3 and 4, the noise reduction device arranges an acceleration sensor 10 on the floor panel 1 to estimate noise in the control space P, and performs noise estimation on the piezoelectric actuator 20 provided on the floor panel 1. A control command signal is generated, and a control sound caused by vibration (wave) of the piezo actuator 20 is input into the vehicle interior. Therefore, the road noise caused by the floor panel 1 is mainly handled as a control target.

  More specifically, as shown in FIG. 3, the excitation force input from the road surface to the tire 2 generates vibration of the floor panel 1, and the vibration of the floor panel 1 is represented by a dotted line in the figure. Is transmitted to the acceleration sensor 10. Thereby, the excitation force of the tire 2 is indirectly detected by the acceleration sensor 10. Moreover, the vibration of the floor panel 1 generates noise in the control space P. For this reason, the excitation force of the tire 2 indirectly becomes noise in the control space P as indicated by a solid line in the figure.

  On the other hand, the piezo actuator 20 generates vibration on the floor panel 1 by operating based on the control command value. The wave generated by the piezo actuator 20 propagates from the piezo actuator 20 to the acceleration sensor 10 on the floor panel, as indicated by a dotted line in the figure. The acceleration sensor 10 detects vibration caused by the wave motion of the actuator 20. And the wave of the piezo actuator 20 becomes noise in the control space P as shown by the one-dot chain line in the figure, as in the case of the excitation force of the tire 2. It is also a component that cancels out noise based on it.

  As shown in FIG. 4, the noise reduction device that reduces the noise in the control space P by the piezo actuator 20 in this way arranges the acceleration sensor 10 and the piezo actuator 20 on the floor panel 1. In this example, there are two control spaces P, Pa and Pb, four acceleration sensors 10 are arranged 10a to 10d, and two piezoelectric actuators 20 are arranged 20a and 20b. When the control space P, the acceleration sensor 10 and the piezo actuator 20 are arranged in this way, each of the acceleration sensors 10a to 10d generates a wave in which the excitation force of all the tires 2 and the vibration caused by the wave of the piezo actuator 20 overlap. Will be detected. In addition, the control spaces Pa and Pb are generated by overlapping the excitation force of all the tires 2 and the sound generated by the wave of the piezo actuator 20.

The setting example of the positions of the acceleration sensor 10, the piezoelectric actuator 20, and the control space P is not limited to the configuration shown in FIG. However, the number of acceleration sensors 10 is generally required to be larger than the number of vibration sources. Specifically, the number and installation positions of the acceleration sensors 10 are sufficiently high in coherency C _xy (ω) between each acceleration sensor 10 and the sound pressure of noise in the control space P as shown in the following Equation 1. To be determined. The coherency C _xy (ω) is preferably 0.9 or more.

Here, P _xy (ω) in Equation 1 is a cross power spectrum between acceleration (signal x) and sound pressure (signal y), and P _xx (ω) and P _yy (ω) are auto values of acceleration and sound pressure, respectively. Represents the power spectrum. P ^H represents a Hermitian transpose matrix of P. Such coherency C _xy (ω) represents the degree of causal relationship between the acceleration of the vibration of the floor panel 1 given by the tire 2 and the sound pressure.

  In addition, a sufficient number of piezoelectric actuators 20 may be attached to appropriate positions of the floor panel 1 of the vehicle body in order to reduce noise in the control space P. For this reason, in this embodiment, the acceleration sensor 10 includes four acceleration sensors 10a, 10b, 10c, and 10d, and the piezoelectric actuator 20 includes two piezoelectric actuators 20a and 20b. Note that the control space P may be set to an appropriate position according to the number, position, etc. of the occupant using, for example, an infrared sensor that detects the occupant position in the passenger compartment.

  As described above, the noise reduction device to which the present invention is applied does not provide a microphone for detecting the noise in the control space P, and based on the sensor signal of the acceleration sensor 10 installed on the floor panel 1, the noise in the control space P is reduced. presume. This noise becomes road noise that gives vibration to the floor panel 1. The reason why the acceleration sensor 10 is provided on the floor panel 1 is that high coherency can be obtained between road noise and vehicle interior noise.

  In addition, since all noises generated from the floor panel 1 are included as control targets, part of engine noise and wind noise generated by air flowing through the bottom of the vehicle body can be handled in the same manner. In addition, the scope of the effect of the present invention is not limited to the category of noise reduction due to vibration of the floor panel 1, but also for vehicle interior noise generation sources generated by the same mechanism such as a dash panel, a windshield, and a roof panel. If the acceleration sensor 10 and the piezo actuator 20 are provided at the part, similarly, noise generated by vibration of the part can be reduced.

  As shown in FIG. 5, the noise reduction device described above reduces noise generated in the control space P of the driver seat or the passenger seat and the control space Pb of the rear seat as the control space P. In addition, the noise reduction device includes a plurality of acceleration sensors 10a, 10b, 10c, and 10d (hereinafter simply referred to as “acceleration sensor 10” when collectively referred to) as vibration detection means for detecting the vibration of the vehicle body. A plurality of piezo-electric actuators (Piezo-electric actuators) 20 which are wave applying means for generating vibrations such as vibration or sound to reduce indoor noise, and the acceleration sensor 10 and the piezo actuators 20 are connected and mounted on a vehicle. And a control device 30. In FIG. 5, the control device 30 is illustrated outside the vehicle for explanation.

  In this embodiment, the wave applying means is described as generating the vibration like the piezo actuator 20, but noise may be reduced by generating sound directly like a speaker. good. Further, the noise reduction apparatus estimates the noise in the control space P from the signal of the acceleration sensor 10 without using a microphone as the vibration detection means.

  The control device 30 includes amplifying units 31a, 31b, 31c, 31d for detecting signal amplification connected to each of the acceleration sensors 10 and amplifying units 31e, 31f for amplifying output signals connected to each of the piezo actuators 20 (hereinafter referred to as the amplifying units 31e, 31f). In general, it is simply referred to as “amplifier 31”.) And a controller 32 that calculates and outputs a control command value for reducing vehicle interior noise. The amplification units 31a to 31d for amplifying the detection signal also have a function of converting between charge and voltage when the acceleration sensor 10 is a so-called charge type.

  When reducing the noise in the control space P, the control device 30 amplifies the detection signal from the acceleration sensor 10 by the amplification unit 31 and inputs the detection signal from the controller 32. Then, the control device 30 estimates noise by the controller 32 using the amplified detection signal, generates a control command value that reduces the noise, and sends the control command value to the piezo actuator 20 via the amplification unit 31. Supply and vibrate floor panel 1.

  As shown in FIG. 6, the controller 32 includes an A / D conversion unit 41, a noise control unit 42, an abnormality detection unit 43, and an amplification unit connected to the amplification units 31a, 31b, 31c, and 31d. A D / A converter 45 connected to 31e and 31f and an arithmetic unit 46 are provided. Note that the noise control unit 42 and the abnormality detection unit 43 in the control device 30 are configured by computer hardware including a CPU, a ROM, a RAM, etc., but in FIG. , Explain. In the present embodiment, the controller 32 is mounted on a so-called digital computer, and the calculation is executed every control cycle of, for example, 1 msec.

The A / D conversion unit 41 converts each sensor signal supplied from the amplification units 31a, 31b, 31c, and 31d into digital acceleration information α ₁ , α ₂ , α ₃ , α ₄ , and a noise control unit 42 and the abnormality detection unit 43.

The noise control unit 42 uses the acceleration information α ₁ , α ₂ , α ₃ , α ₄ supplied from the A / D conversion unit 41 to operate the piezo actuator 20 so as to reduce noise in the control space P. calculating a control command value u ₁ of.

The noise control unit 42 is designed by using, for example, the H2 control method as described below, and calculates the control command value u ₁ based on the acceleration information α ₁ , α ₂ , α ₃ , α ₄ .

That is, the noise control unit 42 includes first to fifth transfer function calculation units 51 to 55 and calculators 56 and 57 as shown in FIG. In the noise control unit 42, the first transfer function calculation unit 51 inputs a vector d having road noise source excitation forces d ₁ , d ₂ , d ₃ , and d ₄ as elements, and a sound pressure vector in the control space P. Is calculated. The second transfer function calculating unit 52 inputs a vector d having road noise source excitation forces d ₁ , d ₂ , d ₃ , and d ₄ as elements, and calculates a vibration (sound pressure) level applied to the acceleration sensor 10. .

The third transfer function calculation unit 53 is an H2 controller and corresponds to the noise control unit 42. The third transfer function calculation unit 53 can be designed to design the noise control unit 42 that calculates the control command value u ₁ that can reduce noise in the control space P. The control command value u ₁ is a vector having control command values u ₁₁ and u ₁₂ for the two piezoelectric actuators 20 as elements. The fourth transfer function calculation unit 54 is a vibration (sound pressure) vector applied from the piezo actuator 20 to the acceleration sensor 10 when the piezo actuator 20 is operated by the control command value u ₁ calculated by the third transfer function calculation unit 53. Is calculated. Therefore, the third transfer function calculation unit 53 adds the vibration (sound pressure) level calculated by the second transfer function calculation unit 52 and the vibration (sound pressure) level calculated by the fourth transfer function calculation unit 54. A vibration (sound pressure) level α is supplied. This α is a vector having acceleration information α ₁ , α ₂ , α ₃ , and α ₄ detected by the acceleration sensor 10 as elements.

The fifth transfer function calculation unit 55 inputs the control command value u ₁ calculated by the third transfer function calculation unit 53, and when the piezo actuator 20 is operated by the control command value u ₁ , the piezo actuator 20 controls the control space P. The vibration (sound pressure) level applied to the is calculated.

The calculator 57 adds the vibration (sound pressure) level that road noise gives to the control space P and the vibration (sound pressure) level given to the control space P by the piezo actuator 20 calculated by the first transfer function calculator 51. Then, the vibration (sound pressure) level SPL1 actually applied to the control space P is calculated. The vibration (sound pressure) level SPL1 becomes two control spaces Pa, vibration of Pb (sound pressure) vector when the piezoelectric actuator 20 is operated by the excitation force and the control command value u ₁ of the tire 2.

In such a design model, acceleration information α ₁ , α ₂ , α ₃ , α ₄ is given to the third transfer function computing unit 53, and d is actually given to give a vibration (sound pressure) level SPL1 in the control space P. The third transfer function calculator 53 of the H2 controller is designed so as to minimize the H2 norm to the H2.

In such a design model, α is defined as in the following equations 2 and 6, d is defined as in the following equation 3, SPL1 is defined as in the following equations 4 and 5, and control is performed. The command value u ₁ is defined as shown in Equation 7 below.

As described above, the noise control unit 42 can calculate the control command value u ₁ that minimizes the noise in the control space P based on the acceleration information α supplied from the A / D conversion unit 41.

The noise control unit 42 is supplied with a signal indicating an abnormal state when an abnormality of the piezo actuator 20 is detected by the abnormality detection unit 43, as will be described in detail later. At this time, the noise control unit 42 newly calculates a control command value u ₁ in order to perform noise control using the piezo actuator 20 that reduces noise control effect in the control space P but at least no abnormality is detected. .

The abnormality detection unit 43 detects an abnormality of the piezo actuator 20 based on the acceleration information α. When the abnormality detection unit 43 detects an abnormality, a signal indicating the abnormal state is supplied to the noise control unit 42. In addition, the abnormality detection unit 43 outputs a control command value u ₂ for generating a test vibration in order to detect an abnormality in the piezo actuator 20. The control command value u ₂ is output only at the timing for determining the abnormality of the piezo actuator 20.

The control command value u ₁ for reducing noise in the control space P output from the noise control unit 42 and the control command value u ₂ for generating test vibration output from the abnormality detection unit 43 are calculated by a calculator 46. And supplied to the D / A converter 45.

The D / A conversion unit 45 converts the sum of the control command value u ₁ output from the noise control unit 42 and the control command value u ₂ output from the abnormality detection unit 43 into an analog signal. The converted analog signals drive the piezo actuators 20a and 20b via the amplifiers 31e and 31f.

  When an abnormality occurs in a certain piezo actuator 20 in the controller 32 of the control device 30 configured as described above, an optimum wave for reducing noise in the control space P cannot be generated by the plurality of piezo actuators 20. . The abnormality of the piezo actuator 20 includes a direct abnormality such as disconnection, breakage, deformation or deterioration of the piezo actuator 20 alone. In addition to this, the abnormality of the piezo actuator 20 includes indirect abnormality such as deterioration of the installation state of the piezo actuator 20 such as deformation of the floor panel 1. Depending on the installation state of the piezo actuator 20, unintentional waves may be generated from the piezo actuator 20, or the piezo actuator 20 may not be able to operate as designed.

  In a situation where the piezo actuator 20 is abnormal, noise control in the control space P cannot be performed efficiently. Further, depending on the type and degree of abnormality of the piezo actuator 20, an inappropriate wave may be generated for the noise control in the control space P, and the noise in the control space P may increase. For this purpose, the control device 30 is provided with an abnormality detection unit 43 for detecting the above of each piezo actuator 20.

The abnormality detection unit 43 includes a test vibration application unit 61 and an abnormality determination unit 62, as shown in FIG. The abnormality detection unit 43 outputs a control command value u ₂ for the test vibration application unit 61 to detect an abnormality of the piezo actuator 20, and the acceleration sensor 10 detects vibration according to the control command value u _2. The abnormality determination unit 62 performs abnormality detection based on the acceleration information α.

Test vibration applying unit 61, the abnormality detecting unit 43 outputs a control command value u ₂ for generating a test vibration sound pressure increase is small in the control space P. The abnormality determination unit 62 detects an abnormality of the piezo actuator 20 from the acceleration information α detected by the control command value u ₂ that generates the test vibration.

The test vibration application unit 61 includes, for example, acceleration information α, frequency characteristics G _f ^SPL (ω) from the piezoelectric actuator 20 to the control space P stored in advance, and frequency characteristics from the piezoelectric actuator 20 to the acceleration sensor 10. A control command value u ₂ for generating a test vibration is calculated from G _f ^α (ω).

The test vibration application unit 61 includes at least one frequency component in the control command value u ₂ for generating the test vibration. In the first embodiment, the test vibration application unit 61 generates a control command value u ₂ that generates a test vibration having a single frequency component.

The test vibration application unit 61 generates a control command value u ₂ for generating a test vibration for each of the piezoelectric actuators 20 a and 20 b and outputs the control command value u ₂ to the computing unit 46. Control command value u ₂ per piezoelectric actuator 20, when the generated the vibration each test by operating the respective piezoelectric actuators 20 by a respective control command value u _2, control space P due to test vibrations The frequency component and the intensity are calculated so as to cancel the noise in the test vibration between the test vibrations.

The test vibration application unit 61 receives a sensor signal from an occupant sensor that detects the presence or absence of an occupant for each seat mounted on the vehicle, and the occupant in the seat where the occupant is detected to be seated is detected. It is desirable to generate a control command value u ₂ that reduces noise generated due to test vibrations in the control space P surrounding the head.

The acceleration information α detected by the acceleration sensor 10 includes, in addition to the acceleration detected by the acceleration sensor 10 due to the operation of the piezo actuator 20 according to the control command value u ₂ for generating the test vibration, the acceleration of the tire 2. The acceleration based on the result of the vibration of the floor panel 1 by the piezo actuator 20 according to the acceleration due to the vibration force and the control command value u ₁ is included. For this reason, it is considered that the acceleration component other than the acceleration generated by the piezo actuator 20 in accordance with the control command value u ₂ that generates the test vibration is noise with respect to the test vibration for detecting an abnormality. In particular, when noise such as traveling of the vehicle is large, the accuracy of detecting an abnormality of the piezo actuator 20 may be reduced.

Therefore, when the plurality of piezo actuators 20 are operated according to the control command value u _2, the noise in the control space P due to each test vibration is sufficiently canceled out, and the acceleration sensor 10 sets the control command value u ₂ . A single frequency of the control command value u ₂ for generating the test vibration is set so as to satisfy any of the following three requirements so that the test vibration based on the noise is detected without being canceled by noise. .

Requirement (1): The frequency of each control command value u ₂ depends on the frequency characteristic G _f ^SPL (ω) from the piezo actuator 20 to the control space P and the frequency characteristic G _f ^α (ω) from the piezo actuator 20 to the acceleration sensor 10. In both cases, the anti-resonance point is not satisfied. Requirement (2): The test vibration is a frequency band in which the gain from the piezo actuator 20 to the acceleration sensor 10 is large. That is, when a test vibration is output by the piezo actuator 20, the vibration detected by the acceleration sensor 10 becomes a predetermined value or more.

Requirement (3): The acceleration component other than the acceleration component by the control command value u ₂ for generating the test vibration is a frequency band. Among these requirements, the requirement (1) and the requirement (2) are the piezoelectric actuator 20 and It is determined at the time of designing to determine the arrangement of the acceleration sensor 10 (determined when installed). Therefore, the test vibration applying unit 61 may store the frequency set F satisfying the requirement (1) and the requirement (2) in the memory in advance.

  The frequencies (frequency set F) satisfying the requirements (1) and (2) are set as follows and stored in the memory of the test vibration applying unit 61 in advance.

FIG. 9 shows an example of the gain characteristic of the frequency characteristic G _f ^SPL (ω) from the piezo actuator 20 to the control space P, and FIG. 10 shows the frequency characteristic G _f ^α (ω) from the piezo actuator 20 to the acceleration sensor 10. An example of a gain characteristic is shown.

As shown in FIG. 9, the anti-resonance in the gain characteristics from all the piezo actuators 20 to all the control spaces P, such as the frequency ω ₁ and the frequency ω _{2 that} are anti-resonance points in the frequency characteristic G _f ^SPL (ω). Let FSPL be the set of frequencies. In addition, as shown in FIG. 10, all acceleration sensors 10 from all piezo actuators 20 have frequencies ω ₃ , ω ₄ , ω ₅ , and ω _{6 that} are anti-resonance points in the frequency characteristic G _f ^α (ω). A set of anti-resonance frequencies in the gain characteristic is Fα.

Further, frequency domain △ .omega.a higher frequency band than the frequency omega ₆ shown in FIG. 10 is a frequency band in which the gain increases from the piezoelectric actuator 20 to the acceleration sensor 10. As in this frequency band, a set of frequencies Fωa having large gain characteristics from all the piezoelectric actuators 20 to all the acceleration sensors 10 is obtained. A set obtained by excluding the anti-resonance point set FSPL and the set Fα from the frequency set Fωa at which a high gain is obtained may be set as the frequency set F satisfying the requirements (1) and (2).

Next, the test vibration requirement (3) that the acceleration component other than the acceleration component based on the control command value u ₂ for generating the test vibration is in a small frequency band is the requirement (1) by the test vibration application unit 61. And the frequency set F satisfying the requirement (2) is selected as a single frequency of the control command value u ₂ for generating the test vibration.

As described above, the test vibration applying unit 61 includes a combination of the piezoelectric actuators 20 that generate the test vibration of the same frequency that satisfies any of the requirements (1), (2), and (3) as described above. The control command value u ₂ having at least one can be generated so that the noise in the control space P is not increased by mutual test vibration.

  In addition, by selecting a single frequency so as to satisfy the requirement (2), the test vibration applying unit 61 causes the vibration detected by the acceleration sensor 10 to exceed a predetermined value when the test vibration is output. A combination of the piezoelectric actuators 20 can be obtained.

  Further, the test vibration applying unit 61 includes the piezo actuator 20 in a combination with any one of the piezo actuators 20 and generates a test vibration by all the piezo actuators 20.

In FIG. 11, when only the control command value u ₁ is applied to the piezo actuator 20 without applying the control command value u ₂ for generating the test vibration during the running, to _one piezo actuator 20. An example of the power spectrum of the obtained acceleration information α is shown. Such frequency characteristics of the acceleration information α are acquired by the frequency analysis by the test vibration applying unit 61. And the vibration application part 61 for a test takes the sum of the power spectrum of the acceleration information (alpha) of all the acceleration sensors 10, for example, the power of the frequency set F which satisfy | fills said requirement (1) and requirement (2) The frequency at which the sum of the spectrum becomes the smallest is set to the frequency of the control command value u ₂ that generates the test vibration.

  As a result, the test vibration application unit 61 ensures that the test vibration is generated by vibrations other than the test vibration without increasing the noise in the control space P even when the test vibrations are generated in the plurality of piezoelectric actuators 20. It can be detected by the acceleration sensor 10.

In addition to setting the frequency of the control command value u ₂ that satisfies the requirements (1), (2), and (3) as described above, a frequency outside the audible range is set as the frequency of the control command value u _2. It is desirable. Thereby, even if the test vibration is generated, the sound pressure does not increase in the control space P. However, in order to generate vibration of the floor panel 1 at a frequency below the audible range, it is necessary to generate a large excitation force by the piezo actuator 20. Further, in order to vibrate the floor panel 1 in an audible frequency range or higher, the vibration transmitted from the piezo actuator 20 to the acceleration sensor 10 is attenuated, and the acceleration sensor 10 may not be able to detect a sufficiently large test vibration. In addition, it is necessary to shorten the sampling period for sampling the acceleration signal of the acceleration sensor 10 into the acceleration information α, and the scene where the frequency outside the audible range can be used is limited, for example, the calculation load increases.

Next, a method of reducing the test vibration generated by the control command value u ₂ that generates the test vibration in the control space P from becoming noise in the control space P will be described.

  For example, if the test vibration is a sine wave signal, one of the piezoelectric actuators 20 that generates the base test vibration is selected from all the piezoelectric actuators 20a and 20b, and the remaining one is the base. A piezo actuator 20 that generates test vibration that reduces the sound pressure in the control space P generated by the test vibration (hereinafter referred to as test noise reduction piezo actuator 20). The base test vibration and the test vibration for reducing the sound pressure due to the test vibration are combined to form the test vibration in the entire noise reduction apparatus. The process of selecting the piezo actuator 20 that generates the test vibration of the base and the piezo actuator 20 for reducing the noise during the test may be set in advance, and is selected by the test vibration applying unit 61 as appropriate. May be.

  Here, the distribution of the piezo actuators 20 that generate the base test vibration and the piezo actuators 20 for noise reduction during testing is not limited to the number of piezo actuators 20. When the number of piezo actuators 20 for noise reduction during testing can be allocated more than the number of control spaces NSPL, the same number of piezo actuators 20 as the number of control spaces NSPL are used as piezo actuators 20 for noise reduction during testing. The remaining piezo actuators 20 are used as piezo actuators 20 that generate test vibrations that are base sine waves. In this embodiment, one piezo actuator 20 that generates base test vibration is used, and the remaining one piezo actuator 20 is used as a piezo actuator 20 for noise reduction during testing.

As shown in FIG. 12, a block diagram of the control system that does not increase the noise in the control space P when the test vibration is generated receives the control command value u ₂₁ for the test vibration as a base. The control command value u ₂₁ is supplied to the transfer function calculation unit (C _t (s)) 71. Based on the control command value u ₂₁ , the transfer function calculation unit (C _t (s)) 71 calculates a control command value u ₂₂ that reduces noise caused by the test vibration as a base. This transfer function calculation unit (C _t (s)) 71 is designed as an H2 controller, like the noise control unit 42. The transfer function calculation unit (G _f2 ^SPL (s)) 73 receives the control command value u _22, and the piezo actuator 20 for noise reduction during the test generates vibration by the control command value u ₂₂ , and the vibration is controlled. A vibration (sound pressure) level applied to the space P is obtained. The transfer function calculation unit (G _f1 ^SPL (s)) 72 receives the control command value u ₂₁ for the test vibration as a base, and calculates the vibration (sound pressure) level that the test vibration gives to the control space P. . The sum of the vibration (sound pressure) level that the base test vibration gives to the control space P and the vibration (sound pressure) level that reduces the base test vibration is calculated by the computing unit 74 and Is output as a vibration (sound pressure) level SPL2 applied to the control space P.

In such a design model, a control command value u ₂₁ for a test vibration as a base is given, and the transmission of the H2 controller so as to minimize the H2 norm to the vibration (sound pressure) level SPL2 in the control space P. A function calculation unit (C _t (s)) 71 is designed. Alternatively, using a map set in advance based on the transfer function calculation unit G _f1 ^SPL (s) 72 and the transfer function calculation unit G _f2 ^SPL (s) 73, the control space P and the sine wave frequency of the test vibration as a base Accordingly, the frequency, amplitude and phase of the control command value for operating the piezo actuator 20 for noise reduction during testing may be set.

  As described above, the sound pressure SPL in the actual control space P is the sum of the vibration (sound pressure) level SPL1 described in FIG. 7 and the vibration (sound pressure) level SPL2 described above. It is possible to output the test vibration while minimizing the increase in sound pressure from the vibration (sound pressure) level SPL1, which is the result of reducing the noise generated based on the vibration of the floor panel 1. In addition, the noise in only the control space P is reduced intensively, and when the number of piezo actuators 20 for noise reduction during the test can be larger than the number of control spaces P, the noise is sufficiently caused by the test vibration. The vibration (sound pressure) level SPL2 can be reduced.

  Further, when the piezo actuator 20 is operated, even when there is no vibration frequency that has little influence on the sound pressure in the control space P, the noise due to the test vibration cancels each other as described above. Increasing the value does not increase the noise in the control space P. Further, since the amplitude of the test vibration can be increased while selecting a frequency with less noise, it becomes easier to detect a change in the acceleration information α of the acceleration sensor 10 caused by the test vibration, and the piezo can be used even during traveling where noise is likely to occur. An abnormality of the actuator 20 can be detected.

<Abnormality judgment of piezo actuator 20>
Next, as described above, the piezo actuator 20 is operated so as to reduce the noise in the control space P based on the excitation force of the tire 2, and at the same time, a test vibration for detecting an abnormality of the piezo actuator 20 is generated. In this case, a process in which the abnormality determination unit 62 determines the abnormality of the piezo actuator 20 will be described.

The abnormality determination unit 62 shown in FIG. 8 calculates an abnormal state of the piezo actuator 20 from the acceleration information α and the frequency characteristic G _f ^α (ω) from the piezo actuator 20 to the acceleration sensor 10 stored in advance.

A frequency component u ₂ (ω) of a control command value u ₂ that generates a test vibration of a certain frequency ω and a frequency component α _f (ω of an acceleration component α _f caused by the control command value u ₂ that generates a test vibration. ) Is expressed by Equation 8 below.

A test vibration is generated based on the control command value u ₂ output by the test vibration application unit 61, and an acceleration component α _f (ω _f resulting from the control command value u ₂ is obtained by reducing noise at the frequency of the test vibration. ) is considered to be substantially equal to the acceleration information alpha (omega _f), using the relationship shown in the equation 8, the acceleration information alpha (omega _f), the control command value u _{2 (ω f} as in equation 9 below) The estimated value u ₂ hat (ω _f ) is calculated.

Then, the abnormality determination unit 62 determines whether the difference between the control command value u ₂ (ω _f ) and the estimated value u ₂ hat (ω _f ) of the control command value u ₂ is a predetermined value or more. 20 is determined to be abnormal.

However, in order to estimate the estimated value u ₂ hat (ω _f ) of the control command value u ₂ using the above equation 9, the following two conditions must be satisfied.

Condition (1): The number of elements of acceleration information α (ω _f ) (the number of acceleration sensors 10 used for abnormality detection) is the number of elements of control command value u ₂ (ω _f ) (piezo actuator 20 that outputs a test vibration). Condition (2): The frequency characteristic G _f ^α (ω _f ) from the piezo actuator 20 to the acceleration sensor 10 is full rank. The noise reduction apparatus in this example includes four acceleration sensors 10, 10 a to 10 _d. Therefore, the number of elements of the acceleration information α is 4, and the number of elements of the control command value u ₂ is 2 because the piezoelectric actuator 20 is composed of two elements 20a and 20b. For this reason, said condition (1) is satisfy | filled. Further, since the condition (2) is determined at the time of design including the combination of the piezo actuator 20 and the acceleration sensor 10 for detecting an abnormality, in addition to the above requirements (1) and (2), the piezo actuator 20 to the acceleration sensor 10 is used. The frequency set F that takes into account the requirement that the frequency characteristics G _f ^α (ω _f ) of the full rank be taken into consideration may be stored in advance in the memory of the test vibration applying unit 61.

  When the noise control unit 42 receives the abnormal state information indicating the result of the abnormality determination unit 62 and there is an abnormality in the piezoelectric actuator 20, for example, the noise control unit 42 stops the control of the piezoelectric actuator 20 or has an abnormality. The control system is reconfigured so as to perform noise control using the other piezoelectric actuators 20 excluding.

  Note that the abnormality detection unit 43 performs the above multiple times without determining at one time in order to reduce false detections for determining that the piezo actuator 20 is abnormal, and detects an abnormality of the same piezo actuator 20 in any of them. In this case, the piezo actuator 20 may be finally determined to be abnormal. Further, as a technique for reducing erroneous determination of abnormality of the piezo actuator 20, the abnormality detection unit 43 changes the frequency of the test vibration over a plurality of times, and detects the test vibration with the acceleration sensor 10. When an abnormality of the piezo actuator 20 is detected based on the frequency, the piezo actuator 20 may be finally determined to be abnormal.

  As described above, according to the noise reduction device according to the first embodiment to which the present invention is applied, a test vibration is generated by at least one piezo actuator 20, and other than the piezo actuator 20 that has generated the test vibration. Since the piezo actuator 20 reduces noise generated in the control space P due to the test vibration, it is possible to avoid increasing noise in the control space P in order to detect an abnormality of the piezo actuator 20.

  In addition, according to the noise reduction device, each seat mounted on the vehicle includes an occupant detection unit that detects the presence or absence of an occupant and surrounds the occupant head in the seat where the occupant is detected to be seated. Since the noise generated due to the vibration for testing in the space is reduced, it is only necessary to suppress the noise in the control space P where the occupant is present, and the degree of freedom of the piezo actuator 20 is widened to reduce the noise at the occupant position. Can be improved.

  Furthermore, according to the noise reduction apparatus, since one or more combinations of the piezo actuators 20 that generate the test vibration of the same frequency are provided, even if the number of the piezo actuators 20 is larger than the acceleration sensor 10, the acceleration sensor 10 A combination of piezo actuators 20 can be set such that the number of piezo actuators 20 is further reduced, and an independent abnormality of the piezo actuators 20 can be detected based on the acceleration information α.

  Furthermore, according to the noise reduction device, the piezo actuator 20 is included in any combination of the piezo actuators 20, and the vibration for testing is generated by all the wave applying means, so that all the piezo actuators 20 are abnormal. Can be detected simultaneously.

  Furthermore, according to the noise reduction apparatus, as described above, when a vibration for testing is output by the piezo actuator 20 by selecting a frequency that satisfies the requirement (2), it is detected by the acceleration sensor 10. Therefore, the degree of freedom of the acceleration sensor 10 can be fully utilized, and the need for increasing the frequency of the test vibration can be eliminated.

  Furthermore, according to the noise reduction device, the test estimated from the vibration detected by the acceleration sensor 10 based on the transfer characteristic for estimating the vibration detected by the acceleration sensor 10 when the piezoelectric actuator 20 generates vibration. Vibration is detected, and when the estimated vibration for the test is different from the actually generated vibration of the piezo actuator 20, it is determined that the piezo actuator 20 is abnormal. Therefore, the vibration detected by the acceleration sensor 10 is detected. Accordingly, it is possible to determine the abnormality of the piezo actuator 20 reliably and individually.

  Furthermore, according to the noise reduction device, the frequency of the vibration level detected by the acceleration sensor 10 when the test vibration is not applied by the piezo actuator 20 is the frequency of the test vibration generated by the piezo actuator 20. Since disturbance vibrations other than the test vibrations included in the vibrations detected by the acceleration sensor 10 can be reduced, an abnormality of the piezo actuator 20 is detected even when vibrations other than the test vibrations enter during driving. Accuracy can be improved.

  Furthermore, according to this noise reduction apparatus, it is possible to always detect an abnormality of the piezo actuator 20 by generating a test vibration by the piezo actuator 20 every predetermined period. Thereby, the process with respect to abnormality of the piezo actuator 20 can be performed rapidly.

[Second Embodiment]
Next, a noise reduction device according to a second embodiment to which the present invention is applied will be described. In addition, about the part similar to 1st Embodiment mentioned above, the detailed description is abbreviate | omitted by attaching | subjecting the same code | symbol.

The noise reduction apparatus according to the second embodiment uses a single frequency as the frequency of the test vibration, and the number of elements of the acceleration information α (ω _f ) is the control command value u ₂ (for generating the test vibration). A case where the number of elements is smaller than ω _f ) is shown. That is, in the noise reduction device according to the second embodiment, the number of acceleration sensors 10 used for detecting an abnormality of the piezo actuator 20 is smaller than the number of piezo actuators 20 that generate test vibrations. In order to make the number of acceleration sensors 10 smaller than the number of piezo actuators 20, the test vibrations generated by the plurality of piezo actuators 20 are respectively set to the same frequency component, amplitude and phase, and the plurality of piezo actuators 20 are combined into one. The piezo actuator 20 may be used.

  As shown in FIG. 13, such a noise reduction device has three acceleration sensors 10, 10 a, 10 b, and 10 c, and five piezoelectric actuators 20, 20 a, 20 b, 20 c, 20 d, and 20 e, and control There are two spaces P, Pa and Pb. Thus, the number of acceleration sensors 10 is set smaller than the number of piezoelectric actuators 20.

  Such a noise reduction device cannot generate a vibration for a test with a single frequency and cannot detect abnormality of all the piezo actuators 20 at the same time.

  Therefore, for example, by utilizing the frequency degree of freedom, the vibrations for testing at a plurality of frequencies are generated by the piezo actuators 20 so that the abnormality of all the piezo actuators 20 is detected at the same time.

  Such a noise reduction device operates as shown in the following (1) and (2).

(1): The number of test vibration frequencies is determined so that the product of the number of test vibration frequencies and the number of acceleration sensors 10 is greater than the number of piezoelectric actuators (2): independent signal vector q And the number of elements of the signal vector q and the coefficient matrix Q so as to satisfy the following condition when the relationship between the control command value u ₂ for generating the test vibration and the signal vector q is represented by u ₂ = Qq The condition of (2) is
(2-1) The number of elements of the signal vector q is smaller than the product of the number of test vibration frequencies and the number of acceleration sensors 10 (conditions for estimating the test vibration from the acceleration information α).
(2-2) Coefficient matrix Q is full rank (conditions that can be estimated by back-calculating test vibration from acceleration information α)
(2-3) The coefficient matrix Q is divided into the same number of rows as the number of piezoelectric actuators 20, and the rank of the matrix Q ′ obtained by connecting these matrices in the column direction is the same as the number of piezoelectric actuators 20 (all at the same time Conditions for detecting an abnormality of the piezo actuator 20)
First, the operation (1) will be described.

  In order to estimate the vibration for testing from the acceleration information α, the number of acceleration information α (degree of freedom) is larger than the number of independent test vibrations (degree of freedom) as described in the first embodiment. There must be. Since the minimum number of vibrations for testing that can simultaneously detect abnormality of all the piezo actuators 20 is the number of piezo actuators 20, at least the number (degree of freedom) of acceleration information α needs to be greater than or equal to the number of piezo actuators 20. is there. Therefore, by using a plurality of frequencies using frequency freedom, the number of acceleration information α (degree of freedom) is increased to the product of the number of test vibration frequencies and the number of acceleration sensors 10. The number of test vibration frequencies is set so that the number (degree of freedom) of acceleration information α is equal to or greater than the number of piezoelectric actuators 20.

In this example, since the number of piezo actuators 20 is 5 and the number of acceleration sensors 10 is 3, the number of test vibration frequencies is ₂ (ω ₁ and ω ₂₎ and the number of acceleration information α (free The degree is 6 (3 (number of acceleration sensors 10) × 2 (number of vibration frequencies for test)). In this way, the number of test vibration frequencies is set to two with respect to the number of acceleration sensors 10. In this case, the test vibration application unit 61 sets a plurality of combinations of the piezo actuators 20 that generate test vibrations having the same frequency, and generates test vibrations having different frequencies for each combination.

  Next, the operation (2) will be described.

Since the number of test vibration frequencies is _two ω ₁ and ω _{2 as in} the above operation (1), the test vibration is 5 (the number of piezoelectric actuators 20) × 2 (the frequency of the test vibration). ) = 10 degrees of freedom. This case will be described below as an example.

First, a matrix G _f ^α (frequency characteristics from the piezo actuator 20 to the acceleration sensor 10) indicating the relationship between the test vibration and the acceleration component will be described. From Equation 8 above, the frequency component u ₂₁ (ω ₁ ) of the control command value u ₂ that generates the test vibration at a certain frequency ω ₁ and the acceleration component α _f generated by the control command value u ₂ that generates the test vibration. The relationship with the frequency component α _f1 (ω ₁ ) is expressed by Equation 10 below.

Similarly, the frequency component u ₂₂ (ω ₂ ) of the control command value u ₂ for generating the test vibration at a certain frequency ω ₂ and the frequency component α of the acceleration component α _f by the control command value u ₂ for generating the test vibration. The relationship with _f2 (ω ₂ ) is expressed by the following formula 11.

Therefore, when Expression 10 and Expression 11 are put together, the following Expressions 12a to 12f are obtained.

As described above, if the transmission characteristics from the piezo actuator 20 to the acceleration sensor 10 are independent at the frequencies ω ₁ and ω ₂ , the frequency characteristics G _f ^α from the piezo actuator 20 to the acceleration sensor 10 are full rank (here Then rank 6). However, since the frequency characteristic G _f ^α is a horizontally long matrix, the control command value u ₂ for generating the test vibration cannot be calculated backward from the acceleration information α. Therefore, a signal vector q having the relationship shown in the following Expression 13 is used.

From the equations 12 and 13, the following equation 14 is obtained.

Therefore, by setting the number of elements of the signal vector q and the coefficient matrix Q so as to satisfy the conditions (2-1) and (2-2) in (2), the piezoelectric sensor 20 to the acceleration sensor 10 are set. The frequency characteristic G _f ^α Q is a vertically long full rank matrix, and the signal vector q can be estimated from the acceleration information α detected by the acceleration sensor 10 by the following equation 15.

Then, from the signal vector q obtained by the above equation 15, using Equation 13 can be estimated control command value u ₂ of the test vibration. Further, when the condition shown in the condition (2-3) of (2) is satisfied, test vibrations are output from all the piezo actuators 20, so that it is possible to determine abnormality of all the piezo actuators 20 at the same time.

  Specifically, a case where the test vibration and the signal vector q correspond one-to-one is shown.

As shown in FIG. 14, there are three piezoelectric actuators 20 that output test vibration of frequency ω ₁ , 20 a, 20 b, and 20 c, and the frequency components of the test vibration output by each of the piezoelectric actuators 20 are u ₂₁₁ , u ₂₁₂ , u ₂₁₃ , the piezoelectric actuator 20 that outputs the test vibration of the frequency ω ₂ is 20 c, 20 d, 20 e, and the frequency components of the test vibration output by each are u ₂₂₃ , u ₂₂₄ , u ₂₂₅ . As described above, the number of piezo actuators 20 is set to be greater than the number of control spaces P at each frequency so that noise due to test vibrations can be canceled.

In this case, since the number of test vibrations is the same as the product of the number of acceleration sensors 10 and the number of test vibration frequencies, the signal vector q is assigned to all the test vibrations. Therefore, the elements of the signal vector q and six Q1 to Q6, the relationship between the control command value u ₂ and the signal vector q of the test vibrating as follows.

u ₂₁₁ = q1
u ₂₁₂ = q2
u ₂₁₃ = q3
u ₂₂₃ = q4
u ₂₂₄ = q5
u ₂₂₅ = q6
At this time, the coefficient matrix Q is expressed as the following Expression 16.

Further, the frequency characteristic G _f ^α Q is expressed by the following Expression 17.

Since this matrix Q satisfies all the conditions shown in (2), the signal vector q can be estimated from the acceleration information α using the above equation 15. The estimated value of the signal vector q corresponds to the estimated value u ₂ hat of the control command value u ₂ . Then, the piezoelectric actuator 20 in which the difference between the control command value u ₂ and the estimated value u ₂ hat is equal to or greater than a predetermined value is determined to be abnormal. Further, when the difference between the control command value u ₂ and the estimated value u ₂ hat is equal to or larger than a predetermined value, the abnormality of the piezo actuator 20 is also detected when the difference between the signal vector q and the estimated value is equal to or larger than the predetermined value. Judgment can be made.

  As a second specific example, the number of test vibrations is greater than the number of signal vectors q.

As shown in FIG. 15, the piezoelectric actuators 20 that output the test vibration of the frequency ω ₁ are 20 a, 20 b, 20 c, and 20 d, and the frequency components of the test vibrations output by the respective piezoelectric actuators 20 are u ₂₁₁ and u _212. , U ₂₁₃ , u ₂₁₄ . Further, the piezoelectric actuators 20 that output the test vibrations having the frequency ω ₂ are 20b, 20c, 20d, and 20e, and the frequency components of the test vibrations that are output by the piezoelectric actuators 20 are u ₂₂₂ , u ₂₂₃ , u ₂₂₄ , u ₂₂₅ . The number of piezo actuators 20 is set to be greater than the number of control spaces P at each frequency so that noise due to vibration for testing can be canceled.

  In addition, in order to make it easy to calculate an input that can cancel the noise caused by the test vibration at each frequency, the same number of test vibrations as the number of control spaces at each frequency are made independent, and (2) The relationship between the test vibration and the signal vector q is set as follows so as to satisfy the conditions shown.

u ₂₁₁ = q1
u ₂₁₂ = q2
u ₂₁₃ = u223 = q3
u ₂₁₄ = u224 = q4
u ₂₂₂ = q5
u ₂₂₅ = q6
At this time, the coefficient matrix Q of the signal vector q is expressed as the following Expression 18.

Further, the frequency characteristic G _f ^α Q is expressed as the following Expression 19.

Since the coefficient matrix Q of Equation 18 satisfies all the conditions shown in (2), the signal vector q can be estimated from the acceleration information α using Equation 15, and the control command value u ₂ can be estimated using Equation 13. The value u ₂ hat can also be calculated. When the difference between the control command value u ₂ and the estimated value u ₂ hat is equal to or greater than a predetermined value, it is determined that the piezo actuator 20 is abnormal. Further, when the difference between the control command value u ₂ and the estimated value u ₂ hat is equal to or greater than a predetermined value, the same signal vector q is assigned to the same piezo actuator 20, so that the difference between the signal vector q and the estimated value is Even if it is greater than or equal to the predetermined value, it can be determined that the piezo actuator 20 is abnormal.

However, as shown in these specific examples, it is not always necessary to have the same number of test vibrations as the number of control spaces P, and in consideration of the relationship between the signal vector q and the control command value u _2. If the signal vector q is set so as to reduce the sound pressure in the control space P, it is possible to cancel the noise in the control space P caused by the test vibration.

Further, since the estimated value u ₂ hat of the control command value u ₂ can be calculated from the signal vector q using Expression 13, it is not necessary to assign the same signal vector q to the same piezo actuator 20.

Note that, as in the first embodiment and the second embodiment described above, the abnormality determination unit 62 of the abnormality detection unit 43 estimates the estimated value u ₂ hat of the control command value u _{2 and} detects the abnormality of the piezo actuator 20. The determination is not limited to the example, and the abnormality of the piezo actuator 20 may be determined based on the change in the acceleration information α detected by the acceleration sensor 10 when the test vibration is changed.

  In addition, the piezoelectric actuator 20 that outputs a plurality of test vibrations at the same time even if the test vibration frequencies are the same without dividing the test vibration frequencies and outputting the test vibrations at the same time as described above. A plurality of combinations whose number is equal to or less than the number of acceleration sensors 10 may be set, and abnormality of the piezo actuator 20 may be sequentially determined for each of these combinations in terms of time.

  As described above, according to the noise reduction device according to the second embodiment to which the present invention is applied, there are a plurality of combinations of the piezo actuators 20 that generate test vibrations having the same frequency, and different frequencies for each combination. Therefore, even if the number of piezo actuators 20 is larger than that of the acceleration sensor 10, the abnormality of the plurality of piezo actuators 20 can be detected at the same time by using the frequency degree of freedom in the test vibrations. Become.

  Also, according to this noise reduction device, the number of vibrations for testing including two or more frequency components is reduced by the piezo actuator 20 rather than the number of acceleration sensors 10, so that the number of piezo actuators 20 is less than the number of acceleration sensors 10. Even when the number is large, it is possible to determine abnormality of the piezo actuator 20 at the same time by effectively utilizing the degree of freedom of the frequency of the test vibration.

  Further, the noise reduction device generates test vibrations with different frequencies by the piezo actuator 20 and compares the abnormality detection results of the piezo actuator 20 based on the vibrations detected by the acceleration sensor 10, thereby comparing the abnormality of the piezo actuator 20. May be detected. Then, the abnormality detection unit 43 may determine the final abnormality of the piezo actuator 20 by obtaining an abnormality detection result obtained by detecting the abnormality of the piezo actuator 20 for each frequency of the test vibration. Thereby, the certainty of detecting an abnormality of the piezo actuator 20 can be increased, and erroneous detection can be reduced.

[Third Embodiment]
Next, a noise reduction device according to a third embodiment to which the present invention is applied will be described. In addition, about the part similar to embodiment mentioned above, the detailed description is abbreviate | omitted by attaching | subjecting the same code | symbol.

In the noise reduction device according to the third embodiment, the abnormality determination unit 62 of the abnormality detection unit 43 calculates the abnormal state of the piezo actuator 20 from the acceleration information α and the frequency characteristic G _f ^α (ω) stored in advance. To do. In particular, the abnormality determination unit 62 calculates the difference obtained by subtracting the acceleration information α when the test vibration is not applied to the floor panel 1 from the acceleration information α when the test vibration is applied to the floor panel 1. Thus, the noise in the acceleration information α is further reduced, and the acceleration component αf caused by the control command value u ₂ for generating the test vibration is accurately extracted.

  However, since the acceleration information α when the test vibration is applied and the acceleration information α when the test vibration is not applied cannot be measured at the same time, the acceleration component αf caused by the test vibration can be accurately obtained. In order to extract, it is desirable to make the noise situation as similar as possible when the test vibration is applied and when it is not applied.

  Therefore, in the noise reduction device according to the third embodiment, for example, the following measures (1) to (4) are performed so that the change in noise between when the test vibration is applied and when it is not applied is reduced. 4).

  Measure (1): When the vehicle is in a steady state, the acceleration information α when the test vibration is applied and the acceleration information α when the test vibration is not applied are obtained. The time when the vehicle is in a steady state is, for example, a case where there is no accelerator, brake or steering operation by the driver, and the vehicle is traveling straight and the vehicle speed is constant.

  Measure (2): Acquire acceleration information α when applying test vibration adjacent to a time zone for acquiring acceleration information α when applying test vibration.

  Measure (3): As shown in FIG. 17, time zones A1 and A2 for acquiring the acceleration information α at the time of applying the test vibration immediately before and immediately after the time zone B for acquiring the acceleration information α at the time of applying the test vibration are set. Provided, if the noise component of the acceleration information α acquired in the time zone A1 and the acceleration information α acquired in the time zone A2 are close or coincide with each other, the acceleration information α acquired in the time zone B between them is also close in noise component Or suppose that they match.

  Measure (4): Acceleration information α is acquired multiple times when the test vibration is applied and when no test vibration is applied, and the test vibration is applied when all of the noise components when the test vibration is applied are close or equal The noise of the acceleration information α at the time and the acceleration information α when the test vibration is not applied are the same.

  The process of accurately extracting the acceleration component αf resulting from the test vibration is expressed as shown in FIG. The flowchart shown in FIG. 16 is executed by the abnormality determination unit 62.

  In step S1, the abnormality determination unit 62 determines whether or not the steady state of the vehicle continues. If the steady state of the vehicle continues, the process proceeds to step S2, and the vehicle is not in the steady state. The extraction of the acceleration component αf caused by the test vibration is stopped. This process corresponds to the above-described measure (1).

  Next, in step S2, the abnormality determination unit 62 acquires acceleration information α when the test vibration is applied and when the test vibration is not applied. For example, as shown in FIG. 17, the acceleration information α when the test vibration is applied in the time zone B is acquired, and the acceleration when the test vibration is applied in the time zones A1 and A2 before and after this time zone B is obtained. Information α is acquired. This process corresponds to the above-mentioned measures (2) and (3).

In the next step S3, the abnormality determination unit 62 extracts the frequency component αf of the frequency ω caused by the test vibration from the acceleration information α. For example, a discrete Fourier transform (FFT) is applied to the acceleration information α. As a result, the acceleration component αf in the time zone A1 is obtained as α (ω) _A1 , the acceleration component αf in the time zone B is obtained as α (ω) _B , and the acceleration component αf in the time zone _A2 is obtained as α (ω) _A2 .

In the next step S4, the abnormality determination unit 62 determines whether or not the acceleration component α (ω) _A1 acquired in step S3 is equal to the acceleration component α (ω) _A2, and determines that they are equal. Advances the process to step S5, and terminates the process if it is determined that they are not equal. As described above, when the acceleration component α (ω) _A1 and the acceleration component α (ω) _A2 are not equal, the process of stopping the extraction of the acceleration component αf corresponds to the above-described measures (2) and (3). To do.

  In the next step S5, the abnormality determination unit 62 determines the number of acquisitions of the acceleration component αf, and determines whether or not the number of acquisitions of the acceleration component αf is the third. Note that the number of acquisitions of the acceleration component αf is not limited to three, but may be one or more. This process corresponds to the above-mentioned measure (4).

In the next step S6, the abnormality determination unit 62 uses the following equation 20 to perform the test six times from the average value aveα (ω) _B of the acceleration component α (ω) _B when the test vibration is applied three times. The average value aveα (ω) _A of the acceleration component α (ω) _A1 and the acceleration component α (ω) _A2 at the time of applying the vibration is subtracted to extract the acceleration component αf (ω), and the process ends.

  As a result, the abnormality determination unit 62 can accurately extract the acceleration component αf caused by the test vibration.

  However, in order to extract the acceleration component αf, it is not necessary to perform all of the above four measures, and by performing any one of the measures, the vibration for testing excluding the acceleration component due to vibrations other than the vibration for testing is removed. The acceleration component αf resulting from can be obtained. Furthermore, the acceleration component αf can be extracted with higher accuracy as the above-described measures are more applied.

After performing such processing, the abnormality determination unit 62 uses the relationship shown in Equation 8 above to generate a test vibration from the extracted acceleration information α _f as shown in Equation 21 below. The estimated value u ₂ hat of the control command value u ₂ is calculated.

Then, the abnormality determination unit 62 determines that the piezo actuator 20 in which the difference between the control command value u ₂ and the estimated value u ₂ hat is equal to or greater than a predetermined value is abnormal.

  As described above, according to the noise reduction device according to the third embodiment to which the present invention is applied, the vibration detected by the acceleration sensor 10 when the vibration for testing is applied by the piezo actuator 20 by the abnormality detection unit 43, and Since the difference from the vibration detected by the acceleration sensor 10 when no test vibration is applied by the piezo actuator 20 is the vibration of the floor panel 1 detected by the acceleration sensor 10, the vibration detected by the acceleration sensor 10 is used. Since disturbance vibrations other than vibrations due to the included test vibrations can be removed, it is possible to improve abnormality detection accuracy when disturbance vibrations occur during traveling.

  Further, according to this noise reduction device, the abnormality detection unit 43 sets a non-application time period in which the piezo actuator 20 does not generate the test vibration immediately before and after the time period in which the piezo actuator 20 generates the test vibration. When the frequency characteristic of the vibration detected by the acceleration sensor 10 in the immediately preceding non-application time zone coincides with the frequency characteristic of the vibration detected by the acceleration sensor 10 in the immediately following non-application time zone, the piezoelectric actuator 20 The acceleration sensor 10 detects the difference between the vibration detected by the acceleration sensor 10 when the test vibration is applied and the vibration detected by the acceleration sensor 10 when the test vibration is not applied by the piezoelectric actuator 20. Because it is a vibration of the floor panel 1 Only detects an abnormality of the piezoelectric actuator 20 by the disturbance change state of small application time zone of the vibration and the non-application time period, it is possible to improve the abnormality detection precision.

  Further, according to this noise reduction apparatus, when the test vibration is applied, the time zone in which the piezo actuator 20 generates the test vibration is adjacent to the non-application time zone in which the piezo actuator 20 does not generate the test vibration. Therefore, it is possible to reduce disturbance change with respect to the test vibration in the application time zone and the non-application time zone of the test vibration, and to improve the abnormality detection accuracy of the piezo actuator 20. Can do.

  Furthermore, according to this noise reduction device, when the vehicle is stopped or when the vehicle is in a steady state, an abnormality of the piezo actuator 20 is detected, so that the test vibration application time zone and the non-application time zone are detected. Since the disturbance to the test vibration can be removed in a state where the disturbance change with respect to the test vibration is small, the abnormality detection accuracy of the piezo actuator 20 can be improved.

  The above-described embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and various modifications can be made depending on the design and the like as long as the technical idea according to the present invention is not deviated from this embodiment. Of course, it is possible to change.

It is a figure explaining the outline of the noise reduction apparatus to which this invention is applied. It is the schematic which shows the structure of a floor panel. It is a figure which shows the transmission path | route etc. of a vibration. It is a figure which shows the example of arrangement | positioning of the acceleration sensor and piezoelectric actuator in the noise reduction apparatus which concerns on 1st Embodiment to which this invention is applied. It is a block diagram explaining the outline of the noise reduction apparatus which concerns on 1st Embodiment to which this invention is applied. It is a block diagram which shows the internal structure of the controller in the noise reduction apparatus which concerns on 1st Embodiment to which this invention is applied. It is a block diagram which shows the structure of the noise control part which concerns on 1st Embodiment to which this invention is applied. It is a block diagram which shows the structure of the abnormality detection part which concerns on 1st Embodiment to which this invention is applied. It is a figure which shows an example of the gain characteristic of the frequency characteristic from a piezoelectric actuator to control space. It is a figure which shows an example of the gain characteristic of the frequency characteristic from a piezoelectric actuator to an acceleration sensor. An example of a power spectrum of acceleration information obtained by a certain acceleration sensor in the case where only a control command value is applied to a piezo actuator without applying a control command value to the piezo actuator while generating a test vibration during traveling. FIG. It is a block diagram of the control system which does not increase the noise of the control space P when the test vibration is generated. It is a figure which shows the example of arrangement | positioning of the acceleration sensor and piezoelectric actuator in the noise reduction apparatus which concerns on 2nd Embodiment to which this invention is applied. It is a figure which shows the relationship between a piezoelectric actuator and the frequency of the vibration for a test. It is a figure which shows the other relationship between the piezoelectric actuator and the frequency of the vibration for a test. It is a flowchart which shows the process which extracts accurately the acceleration component resulting from the vibration for a test. It is a figure explaining providing the time zones A1 and A2 which acquire the acceleration information at the time of the test vibration application immediately before and after the time zone B for acquiring the acceleration information at the time of the test vibration application.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Floor panel 2 Tire 10a, 10b, 10c, 10d Acceleration sensor 20a, 20b Piezo actuator 30 Control apparatus 31a, 31b, 31c, 31d, 31e, 31f Amplifying part 32 Controller 41 Conversion part 42 Noise control part 43 Abnormality detection part 45 D / A conversion unit 46 computing units 51-55 first to fifth transfer function computing unit,
56, 57 Calculator 61 Test vibration application unit 62 Abnormality determination unit 71 Transfer function calculation unit (C _t (s))
72 Transfer Function Calculation Unit (G _f1 ^SPL (s))
73 Transfer Function Calculation Unit (G _f2 ^SPL (s))
74 Calculator 101 Axle 102 Suspension 103 Member

Claims (18)

A noise reduction device for reducing noise in a predetermined control space in a vehicle interior,
A plurality of vibration detecting means arranged on the vehicle structure for detecting the vibration of the vehicle structure;
A plurality of wave applying means for applying waves to the vehicle structure so as to reduce noise generated inside the vehicle;
An abnormality detecting means for detecting an abnormality of the wave applying means based on the vibration of the vehicle structure detected by the vibration detecting means when the vibration for testing is generated by the wave applying means;
Control means for operating the wave applying means to reduce noise in a predetermined space in accordance with the vibration detected by the plurality of vibration detecting means when the abnormality detecting means does not detect an abnormality of the wave applying means. And
The abnormality detecting means generates a test vibration by at least one wave applying means, and is generated in the predetermined space by the wave applying means other than the wave applying means that has generated the test vibration. Noise reduction device characterized by reducing noise to be performed. Equipped with an occupant detection means for detecting the presence or absence of an occupant for each seat mounted on the vehicle,
The control means reduces noise generated due to the test vibration in a space surrounding an occupant head in a seat detected by the occupant detection means. Item 2. The noise reduction device according to Item 1.   The noise reduction device according to claim 1, wherein the abnormality detection unit includes one or more combinations of wave application units that generate test vibrations having the same frequency.   The said abnormality detection means has two or more combinations of the wave application means which generate the vibration for a test with the same frequency, and generates the vibration for a test with a different frequency for each combination. Noise reduction device.   The abnormality detection means includes the wave application means in a combination of any of the wave application means, and generates vibration for testing by all the wave application means. The noise reduction device described.   The abnormality detecting means is a combination of wave applying means so that the vibration detected by the vibration detecting means becomes a predetermined value or more when the vibration for testing is output by the wave applying means. The noise reduction device according to any one of claims 3 to 5.   The abnormality detecting means is for testing estimated from the vibration detected by the vibration detecting means, based on a transfer characteristic for estimating the vibration detected by the vibration detecting means when the vibration is generated by the wave applying means. The vibration is obtained, and when the estimated vibration for the test is different from the actually generated vibration of the wave applying means, it is determined that the wave applying means is abnormal. Item 7. The noise reduction device according to any one of Items 6 to 6.   The abnormality detecting means sets a frequency with a small vibration level detected by the vibration detecting means when the test vibration is not applied by the wave applying means as a frequency of the test vibration generated by the wave applying means. The noise reduction device according to any one of claims 1 to 7, wherein   The noise reduction device according to claim 7, wherein the abnormality detection unit reduces the number of test vibrations including two or more frequency components by the wave application unit, rather than the number of the vibration detection units.   The abnormality detecting means sets the number of test vibrations to 1 by making the frequency, amplitude and phase of the test vibrations generated by the plurality of wave applying means the same, and more than the number of vibration detecting means, The noise reduction apparatus according to claim 9, wherein the number of vibrations for testing including two or more frequency components is reduced by the wave applying means.   The abnormality detecting means generates test vibrations having different frequencies by the wave applying means, compares the abnormality detection results of the wave applying means based on the vibrations detected by the vibration detecting means, The noise reduction device according to any one of claims 1 to 10, wherein an abnormality is detected.   The anomaly detection means is detected by the vibration detection means when the test vibration is applied by the wave application means and when the test vibration is not applied by the wave application means. The noise reduction device according to any one of claims 1 to 11, wherein the difference from the detected vibration is the vibration of the vehicle structure detected by the vibration detection means.   The abnormality detection means sets a non-application time zone in which the vibration application means does not generate a test vibration immediately before and immediately after a time period in which the vibration application means generates the test vibration, and the non-application time immediately before the time When the frequency characteristic of the vibration detected by the vibration detecting means in the band and the frequency characteristic of the vibration detected by the vibration detecting means in the non-application time period immediately after the band coincide with each other, the test vibration is applied by the wave applying means. The vehicle structure in which the difference between the vibration detected by the vibration detecting means and the vibration detected by the vibration detecting means when no test vibration is applied by the wave applying means is detected by the vibration detecting means. The noise reduction device according to any one of claims 1 to 11, wherein the noise reduction device is body vibration.   13. The abnormality detection unit makes a time zone in which the vibration for testing is generated by the wave applying unit and a non-application time zone in which the vibration for testing is not generated by the wave applying unit are adjacent to each other. Item 14. The noise reduction device according to Item 13.   15. The abnormality detection unit according to claim 12, wherein the abnormality detection unit detects an abnormality of the wave application unit when the vehicle is stopped or when the vehicle is in a steady state. The noise reduction device described.   16. The abnormality detecting unit detects an abnormality of the wave applying unit over a plurality of times, and determines that the wave applying unit in which the abnormality is detected is abnormal. The noise reduction device according to any one of the above.   16. The abnormality detecting unit according to claim 1, wherein the abnormality detecting unit detects an abnormality of the wave applying unit at all times by generating a test vibration by the wave applying unit every predetermined period. The noise reduction device according to claim 1. A noise reduction method for reducing noise in a predetermined control space in a vehicle interior,
Generating a test vibration by a wave applying means for applying a wave to the vehicle structure;
Detecting the vibration of the vehicle structure by the vibration detecting means when the wave applying means generates a test vibration; and
Detecting an abnormality of the wave applying means based on the detected vibration of the vehicle structure,
In the step of generating the test vibration, noise generated in the predetermined space due to the test vibration is reduced by a wave applying unit other than the wave applying unit that generated the test vibration. Noise reduction method.

Images