JP2010184586A - Device and method for reducing noise - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for reducing a noise, performing a noise reducing control according to a state change of a structure without including sensors such as a seating sensor and a temperature sensor. <P>SOLUTION: A control command signal generating filter 335 generates a control command signal, that is, a driving signal of a piezoactuator for impressing wave motions from an acceleration detection signal on the structure. A transmission characteristic identifying part 331 identifies a transmission characteristic of the structure from a test signal to the acceleration detection signal when the test signal is inputted. A state estimating part 332 estimates the state of a vehicle body such as the number of occupants and a getting-on position, according to the transmission characteristic. A characteristic changing part 333 changes a characteristic of the control command signal generating filter 335 according to the estimated state. A test signal generating part 334 generates the test signal. An adding part 336 adds the control command signal and the test signal. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a noise reduction apparatus and method for reducing noise and vibration by applying a wave that cancels vibration and noise of a structure such as a vehicle body.

  2. Description of the Related Art An active noise suppression device that actively reduces noise by generating a sound that cancels noise with respect to noise generated by a structure such as a vehicle is known. For example, a seating sensor, an infrared camera, or a temperature sensor detects the structural environment such as the occupant position and the passenger compartment temperature, optimizes the controller characteristics according to changes in the environment, and transmits from the actuator to the control space sound pressure. There have been proposed noise reduction devices and noise reduction methods that do not reduce the noise suppression effect even if the characteristics change (for example, Patent Documents 1 and 2).

JP 2007-296886 A JP 2007-296887 A

  However, in the above conventional noise reduction device, it is necessary to include sensors such as a seating sensor, an infrared camera, and a temperature sensor in order to detect a change in the structure environment. There was a problem of becoming higher.

  In order to solve the above-mentioned problems, the present invention provides a vibration detection means for detecting vibration of a structure, a wave application means for applying a wave to the structure, and a drive signal corresponding to the detection signal of the vibration detection means. And a control means for reducing at least one of vibration and noise of the structure to identify a transfer characteristic from the test signal input to the wave applying means to the detection signal. The state of the structure is estimated according to the transfer characteristic, and the characteristic of the drive signal is changed according to the estimated state.

  According to the present invention, it is possible to change the characteristics of the drive signal in accordance with a change in the state of the structure without providing sensors for detecting the state of the structure, thereby increasing the cost without complicating the apparatus. Without lowering the noise reduction performance due to a change in the state of the structure.

It is a figure which shows the main propagation path | route of the vibration of a vehicle body by the influence of the unevenness | corrugation of a road surface, and road noise. It is a figure showing the change of the frequency characteristic of a vibration characteristic and a sound vibration characteristic accompanying a passenger | crew position change. It is a figure showing the change of the frequency characteristic of a vibration characteristic and a sound vibration characteristic accompanying a temperature change. It is a block diagram which shows the structure of Example 1 of the noise reduction apparatus which concerns on this invention. FIG. 3 is a block diagram of a control command signal calculation unit according to the first embodiment. It is a block diagram which shows the noise control system of Example 1. FIG. It is a figure which shows the test signal generation frequency of Example 1. FIG. 3 is a flowchart illustrating a processing procedure of a control command signal calculation unit according to the first embodiment. It is a figure which shows the effect of Example 1. FIG. It is a block diagram which shows the noise control system resulting from the test signal of Example 2. 6 is a flowchart illustrating a processing procedure of a control command signal calculation unit according to the second embodiment. FIG. 6 is a block diagram of a noise control system according to a third embodiment. It is a figure explaining the definition of the number of sensors, actuators, and control spaces based on Example 3. It is a block diagram which shows the structure of the control command signal calculation part of Example 3. It is explanatory drawing of the test signal generation frequency (omega) of Example 3. FIG. 10 is a generalized plant for controller design according to a third embodiment. FIG. 9 is a controller structure diagram of Embodiment 3. It is a flowchart of the calculation which the control command signal calculation part of Example 3 performs.

  Next, embodiments of the present invention will be described in detail with reference to the drawings. Each embodiment described below describes a noise reduction device and method for reducing at least one of vibration and noise inside a vehicle as a structure.

  Prior to the description of the embodiment, vehicle interior noise will be described. Typical causes of vehicle interior noise are engine noise caused by engine vibration, noise caused by intrusion of road surface irregularities from tires during driving (hereinafter referred to as road noise), and driving There is a wind noise generated by the air current. Road noise and wind noise are also present in electric vehicles without engine noise, and the present invention can be applied to electric vehicles.

  FIG. 1 is a conceptual diagram showing main propagation paths of vehicle body vibration and road noise due to the influence of road surface unevenness. This embodiment mainly deals with reduction of road noise.

  The vibration of the tire 2 due to the unevenness of the road surface first enters a highly rigid beam-like member called the member 140 from the mounting portion (not shown) of the axle 120 and the suspension 130. Thereafter, vibration propagates to a plate-like member called floor panel 3 surrounded by the member 140 and having relatively low rigidity, and the floor panel 3 vibrates. And the vibration of the floor panel 3 causes air vibrations in the passenger compartment, causing a resonance phenomenon in the passenger compartment. For this reason, road noise is heard in a predetermined space (hereinafter referred to as control space 6) in the passenger compartment. Although noise is generated by vibration of the roof panel and window glass (both not shown) in addition to the floor panel 3, road noise that mainly enters from the attachment portion of the suspension 130 is caused by vibration of the floor panel 3. I know the relationship is high.

  Therefore, if the vibration of the floor panel 3 is detected by the acceleration sensors 4a to 4d and a wave is applied using the piezo-electric actuators 5a to 5d so as to cancel the vibration, the road noise in the control space 6 is obtained. Can be reduced.

  However, when the piezo actuators 5a to 5d give a wave to the floor panel 3, it is controlled from the actuator input according to the position of the occupant and the temperature of the vehicle body and the interior space as shown in FIGS. 2 (a) and 3 (a). The transfer characteristic up to the sound pressure in the space 6 changes. For this reason, since the control sound deviates from the desired control sound in the control space 6, an error occurs in the noise control, and the noise reduction effect in the control space 6 is deteriorated, which may cause discomfort to the passenger.

  Here, as shown in FIGS. 2 (b) and 3 (b), the transfer characteristics from the actuator input that applies the wave to the floor panel to the sensor output that detects the floor vibration are also the occupant position, the vehicle body temperature, and the cabin temperature. It turns out that it changes according to. Therefore, in this embodiment, the occupant position and temperature are estimated based on the detection signals of the acceleration sensors 4a to 4d when the test signals are input to the piezo actuators 5a to 5d, and noise control is performed based on the estimation results. The controller that calculates the actuator input is corrected. Thereby, without using a microphone, an occupant position detection sensor, or a temperature sensor, it is possible to suppress a reduction in noise reduction effect due to changes in the occupant position or temperature, and to reduce discomfort for the occupants in the vicinity of the control space 6.

  As shown in FIG. 1, in this embodiment, the acceleration sensors are 4a, 4b, 4c, and 4d, the piezoelectric actuators that are wave application units are 4a, 5a, 5b, 5c, and 5d, and the control space is 6a. , 6b is assumed.

  As described above, when there are a plurality of tires as vibration sources, and there are a plurality of acceleration sensors, a plurality of wave application units, and a plurality of control spaces, in each acceleration sensor, the excitation force of all tires and all the wave application units It is detected by overlapping the vibration caused by. Moreover, in each control space, all the tire exciting forces overlap with the sounds generated by all the wave application units.

  Note that the number of acceleration sensors, wave application units, and control spaces is not limited to the number shown in FIG. 1, but may be any number that satisfies the following conditions.

In general, the number of acceleration sensors is required to be larger than the number of vibration sources. If the number and installation positions of the acceleration sensors on the floor panel 3 are determined so that the coherence between the detection signal of each acceleration sensor 4 and the sound pressure of the noise in the control space 6 is sufficiently high (for example, 0.9 or more). Good. Here, the coherence Cxy (ω) indicating the correlation between the signal x (t) and the signal y (t), which is two time functions, is defined by the following equation (1). If the two signals x (t) and y (t) are uncorrelated, the value of Cxy (ω) is 0, and if they match, the value of Cxy (ω) is 1.

Here, Pxy (ω) is the cross power spectrum between the signal x (t) and the signal y (t), Pxx (ω) is the auto power spectrum of the signal x (t), and Pyy (ω) is the signal y ( It is an auto power spectrum of t). P ^H represents a Hermitian transpose matrix of P.

In addition, a sufficient number of piezoelectric actuators 5 may be attached to an appropriate location on the floor panel 3 in order to reduce noise in the control space 6. For example, the control space 6 may be set in a predetermined space in advance, or may be set appropriately in accordance with an occupant position detected by an occupant position detection method described later.
Further, 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 3, but for example, noise in the passenger compartment generated by the same mechanism such as a dash panel, a front glass, and a roof panel (both not shown). The same effect can be obtained for the generation source if the present invention is used for the site.

  Next, Example 1 of the present invention will be described. Example 1 is an example in which a test signal having a plurality of frequencies is applied to one wave application unit to identify transfer characteristics from a test signal to a vibration detection signal.

  FIG. 4 is an overall configuration diagram of the noise reduction device according to the first embodiment. The noise reduction device includes an acceleration sensor 4 (corresponding to the acceleration sensors 4a to 4d in FIG. 1) that detects vibration of the floor panel 3 of the vehicle body 1, and a piezo actuator that is a wave application unit that applies mechanical waves to the floor panel 3. 5 (corresponding to the piezoelectric actuators 5a to 5d in FIG. 1) and a control command signal to be supplied to the piezoelectric actuator 5 according to the detection signal of the acceleration sensor 4, and a wave is applied from the piezoelectric actuator 5 to the floor panel 3. Thus, a control unit 30 that reduces at least one of vibration and noise of the floor panel 3 is provided.

  In this embodiment, road noise entering the control spaces 6a and 6b of the vehicle body 1 from the tire 2 is reduced. The control space 6a is a space that includes the position of the ear of the front seat occupant of the vehicle body 1, and the control space 6b is a space that includes the position of the ear of the rear seat occupant of the vehicle body 1.

  The acceleration sensor 4 is a sensor that detects the acceleration applied to itself. The acceleration sensor 4 is fixed to the lower surface of the floor panel 3 to detect the acceleration, thereby detecting the vibration of the floor panel 3 from the detected acceleration change. can do.

  The piezo actuator 5 is an actuator that uses a piezo effect (piezoelectric effect) in which distortion is proportional to the electric field when an electric field is applied to the piezoelectric element. The piezo actuator 5 is fixed to the lower surface of the floor panel 3, and receives a control command signal that is an electrical signal, and is distorted by itself to apply a control wave to the floor panel 3.

  In this embodiment, a piezoelectric actuator is used as a device for applying a control wave. However, in addition to this, a device capable of changing an electrical signal such as a magnetostrictive transducer into a mechanical wave, There is no particular limitation.

  The control unit 30 includes an amplification unit 31 that amplifies the detection signal of the acceleration sensor 4, an A / D conversion unit 32 that performs analog / digital conversion on the amplified acceleration signal, and a digital control command based on the acceleration signal that has been digitally converted. A control command signal calculation unit 33 that generates a signal, a D / A conversion unit 34 that converts the digital control command signal into an analog signal, and an amplification unit 35 that amplifies the analog control command signal and supplies the amplified signal to the piezoelectric actuator 5. ing.

  The amplifying unit 31 is actually provided with four (or four channels) amplifiers according to the number of acceleration sensors 4a to 4d. Similarly, the amplifying unit 35 includes the number of piezo actuators 5a to 5d. Four (or four channel) amplifiers are provided. In addition, when the acceleration sensor 4 is a charge charge type, the amplification unit 31 also performs conversion between charge and voltage.

  FIG. 5 is a block diagram illustrating a functional configuration of the control command signal calculation unit 33 in the control unit 30. The control command signal calculation unit 33 actually includes, for example, a digital signal processor (DSP), a central processing unit (CPU) that controls the DSP, a program ROM that stores a program of the CPU, a nonvolatile memory, a work It is comprised with the microcomputer provided with RAM for operation.

  The control command signal calculation unit 33 includes a control command signal generation filter 335 that is a digital adaptive filter that generates a control command signal from an acceleration detection signal, and a test signal to an acceleration detection signal when the test signal is input to the piezoelectric actuator. A transfer characteristic identifying unit 331 for identifying the transfer characteristic, a state estimation unit 332 for estimating the state of the vehicle body, for example, the number of passengers and the boarding position, according to the transfer characteristic, and a control command signal according to the estimated state A characteristic changing unit 333 that changes the characteristic of a control command signal that is a drive signal of the piezoelectric actuator 5 by changing the characteristic of the generation filter 335, a test signal generating unit 334 that generates a test signal, and an adding unit 336 are provided. .

  Next, each component of the control command signal calculation unit 33 will be described in detail. First, the control command signal generation filter 335 will be described. The control command signal generation filter 335 calculates the control command signal u from the acceleration detection signal α detected by the acceleration sensor 4. The filter used in the control command signal generation filter 335 may be designed as follows, for example.

  FIG. 6 is a block diagram for calculating the control command signal generation filter 335. A block 3311 (denoted as C in FIG. 6) indicates a filter that is actually mounted on the control command signal generation filter 335. A block 3312 (denoted as Gp-SPL in FIG. 6) indicates a transfer function from the piezoelectric actuator 5 to the control space sound pressure (SPL). Block 3313 (denoted as Gα-SPL in FIG. 6) shows a transfer function from the acceleration detection signal to the control space sound pressure. A block 3314 (indicated as Wc in FIG. 6) represents a weighting function for outputs to the piezo actuators 5a to 5d which are wave application units.

  The transfer function of the block 3312 is obtained by inputting an impulse signal, white noise or the like to the piezo actuator 5 and performing an experiment to measure the sound pressure at that time using a microphone temporarily installed in the control space 6. This can be obtained by system identification using the output signal. For system identification, for example, a function “n4sid” in a tool box “System Identification Toolbox” of a general numerical analysis program MATLAB may be used. In the function “n4sid”, the state space model is identified from the time calendar data using the subspace identification method.

  For the transfer function of block 3313, a microphone is temporarily installed in the control space 6, and an experiment is performed to measure an acceleration detection signal from the acceleration sensor 4 and a control space sound pressure when the vehicle is running. do it. For example, if there are 5 passenger cars, there are 5 seats, and there are 32 different riding patterns depending on whether there are passengers in each seat. The above experiment is performed for these 32 patterns. As a result, 32 types of system identification are performed. If the driver is always on board among the 5 passengers, there are 16 boarding patterns depending on whether there are passengers in the remaining 4 seats.

  Block 3314 sets the control command signal so that it does not exceed the maximum allowable voltage of the piezoelectric actuator 5. At this time, the filter C is designed so as to minimize the transfer function from the acceleration detection signal to the control space sound pressure (SPL) and the norm of the transfer function from the acceleration signal to the control command signal. The filter C may be designed using, for example, the design method of the H2 controller in the tool box “Robust Control Toolbox” of the general-purpose numerical analysis program MATLAB.

  Next, the transfer characteristic identification unit 331 will be described. The transfer characteristic identification unit 331 identifies transfer characteristics from the test signal to the acceleration detection signal based on the acceleration signal detected by the acceleration sensor 4 when the test signal is applied to the piezo actuator 5 as a control command signal.

  Next, the state estimation unit 332 will be described. The state estimation unit 332 estimates the number of passengers and the boarding position as the state of the vehicle body 1 based on the transfer characteristics identified by the transfer characteristic identification unit 331. This estimation determines whether the transfer characteristic is closest to the previously stored 32 or 16 passenger numbers and boarding positions in the case of a passenger car with a capacity of 5 persons. In the following description, the number of passengers and the boarding position are simply referred to as a boarding position or a combination of boarding positions. In addition, as the state of the vehicle, an occupant posture such as whether the occupant of each seat is an adult or a child and whether the occupant is sitting in a good posture or throwing out a foot may be added.

  Then, the state estimating unit 332 instructs the characteristic changing unit 333 to obtain a control command signal characteristic corresponding to the estimated boarding position. The characteristic changing unit 333 changes the parameter of the control command signal generation filter 335 according to an instruction from the state estimating unit 332.

  The test signal generator 334 generates a signal to be input to the piezo actuator 5 in order to identify the transfer characteristic from the input of the piezo actuator 5 to the acceleration detection signal.

The test signal generator 334 applies a test signal to any one of the piezo actuators 5 and sets the signal applied to the other piezo actuators 5 to 0. A sine wave signal having a plurality of frequencies is used as the test signal. The frequency of the sine wave signal to be applied may be continuous, and the test signal may be applied to a certain frequency band. By using such a test signal, the state estimation unit 332 compares the transfer characteristics, and when the occupant position combinations having close transfer characteristics are specified, the frequency band to be compared increases, so the occupant position is detected more accurately. become able to.
Here, the test signal does not necessarily include a plurality of frequencies at the same time, and a sine wave signal having a single frequency may be input while shifting the time. In this case, the transfer characteristic identification unit 331 identifies the transfer characteristic after the test signal has input all of a plurality of frequencies. Whether the test signal input is completed may be determined by applying the end signal to the test signal at the end of the test signal input and detecting it by the transfer characteristic identification unit 331.

  The frequency of the test signal may be set to a frequency at which the sound characteristic gain from the piezo actuator to the control space sound pressure is low as shown in FIG. By applying the test signal to such a frequency, the test signal is less likely to be a sound, so that noise caused by the test signal can be reduced.

  When the vehicle is an internal combustion engine vehicle, low frequency sound outside the audible band may be input to the piezo actuator 5 as a test signal between the time of key insertion and the engine start. Since there is no vibration from the road surface or engine from when the key is inserted until the engine is started, the sensor vibrates even in a frequency band where the piezo actuator cannot output a sufficient wave, such as a frequency outside the audible band. Can be detected.

  In addition, since the test signal cannot be heard by the occupant, noise caused by the test signal can be completely suppressed. Although it is conceivable that the occupant position changes after the engine is started, when the occupant position is estimated again by the state estimation unit 332 thereafter, only the transfer characteristics of the nearby occupant position combinations are estimated based on the occupant position estimated at this time. As a result, the time required for occupant position estimation can be shortened.

In addition, when re-estimating the occupant position, the number of transfer characteristics up to the control space sound pressure to be considered can be limited by selecting the frequency of the test signal based on the occupant position estimated at this time. Thus, it becomes possible to select a frequency band that is less prone to noise at the passenger position.
The test signal may be applied by selecting one of the piezo actuators 5a to 5d whose transfer characteristics from the piezo actuator 5 to the acceleration sensor 4 are likely to vary depending on the occupant position. When the state estimation unit 332 compares the transmission characteristics and identifies the occupant position combinations having similar transmission characteristics, if the difference in the transmission characteristics depending on the occupant positions is large, the transmission characteristics of different occupant position combinations are easily distinguished. This makes it easier to identify the occupant position with higher accuracy.

  In addition, the acceleration detection signal detected after applying the test signal is higher than the acceleration detection signal detected by the acceleration sensor 4 before applying the test signal to the piezo actuator 5 at the frequency at which the test signal is applied. It is better to adjust the input of the test signal so that becomes larger. By increasing the acceleration detection by the test signal, the SN ratio of the acceleration detection signal when the test signal is applied is improved, and the identification accuracy of the transfer characteristic is improved.

  In addition, by applying the test signal when the vehicle is stopped or at an extremely low speed, the input from the road surface does not enter, so the SN ratio of the acceleration detection signal when applying the test signal is improved, and the identification accuracy of the transfer characteristic is improved. .

  The adder 336 adds the signal applied to the piezo actuator 5 calculated by the control command signal generation filter 335 and the test signal from the test signal generator 334 to generate a control command signal to be applied to the piezo actuator 5.

  Next, the process of the control command signal calculation unit 33 will be described with reference to the flowchart of FIG. In step S101, the transfer characteristic identification unit 331 checks whether the test signal is applied by the test signal generation unit 334. When the test signal is applied, the piezoelectric actuator 5 to which the test signal is applied and the frequency thereof are specified, and the process proceeds to S102. When the test signal is not applied, the characteristic changing unit 333 sets the control command signal generation filter 335 to the same characteristic as the previous time and ends the process.

  In step S102, the transfer characteristic identifying unit 331 identifies the transfer characteristic from the test signal applied to the piezoelectric actuator to the acceleration detection signal detected by the acceleration sensor 4. Actually, since the input calculated by the control command signal generation filter 335 is also applied to the piezo actuator 5 at the same time as the test signal in order to control noise, the test signal and the output of the control command signal generation filter 335 are added. The characteristics from the control command signal added by the unit 336 to the acceleration detection signal are identified. However, since the transmission path is the same, the identification result of the transmission characteristic coincides with the case where only the test signal is applied.

  For identification of this transfer characteristic, a non-parametric system identification method such as “weighted overlapped segment averaging method” may be used. The system identification may be limited to only the piezoelectric actuator to which the test signal specified by the transfer characteristic identification unit 331 is applied and the frequency to which the test signal is applied. By limiting the frequency to be identified, the time required for the calculation can be shortened.

  In step S103, the state estimation unit 332 compares the transfer characteristic for each occupant position combination stored in advance in the memory in the state estimation unit 332 with the result of system identification in step S102. The transmission characteristics for each occupant position combination stored in the memory include an impulse signal and white noise applied to the piezo actuator 5 for every possible number of occupants and occupant positions, and acceleration obtained by the acceleration sensor 4 obtained at that time. It can be obtained by performing system identification using the detection signal. For system identification, a method similar to that used when applying a test signal may be used.

  In step S104, the state estimation unit 332 determines whether there is an occupant position combination having a transfer characteristic close to the result of system identification by the transfer characteristic identification unit 331. If there is a combination with similar characteristics, the process moves to step S105. If there is no combination with similar characteristics, the process proceeds to step S106. Here, for example, the determination as to whether there is an occupant position combination with similar transfer characteristics may be performed as follows. The difference between the current transfer characteristic identified in step S102 and the transfer characteristic at each occupant position combination is calculated for each frequency. Then, the occupant position combination with the smallest sum of absolute values of the differences in the transfer characteristics is set as the occupant position at the current time. However, if any of the gain and phase of the difference in transfer characteristics at each frequency exceeds the threshold values Tg and Tp, it is determined that there is no passenger position combination with similar transfer characteristics. The threshold values Tg and Tp may be set to values that can provide a desired control effect even when the characteristics are shifted by Tg and Tp, respectively, from the transfer characteristics at the time of filter design when performing noise control.

  In step S105, the characteristic changing unit 333 switches from the filters stored in the memory in advance to the control command signal generation filter 335 corresponding to the estimated occupant position combination, and ends the process. At this time, the control command signal generation filter 335 to be switched may be designed so that the noise reduction effect becomes higher in the space where the passenger is present. By using such a control command signal generation filter, control can be concentrated in a specific space, so that more effective noise reduction can be realized.

  In step S106, it is determined that there is a factor that causes the transfer characteristic to fluctuate other than the difference in the occupant position, and the characteristic changing unit 333 ends the process without switching the control command signal generation filter 335.

  The effect at the time of using the noise reduction apparatus described above is shown. FIG. 9 shows the noise level in the control space 6 after the noise reduction control is performed. When the occupant position is specified according to the embodiment, since an appropriate control command signal generation filter can be selected, a high noise reduction effect is obtained, but the occupant is not used without using the embodiment. If the position cannot be specified, it can be seen that the noise reduction effect is reduced.

  Here, in the present embodiment, the method of detecting the occupant position as the state of the vehicle body 1 has been described. However, not only the transfer characteristics for each occupant position combination but also each occupant position combination is stored in the memory in the state estimation unit 332. On the other hand, transmission characteristics can also be stored for each constant temperature (for example, in increments of 5 ° C.) in the vehicle body or the passenger compartment. Then, the transfer characteristic closest to the current characteristic identified by the transfer characteristic identification unit 331 may be selected. And the characteristic change part 333 should just change the characteristic of the control command signal generation filter 335 so that it may become the control command signal generation filter characteristic corresponding to the passenger | crew position and temperature.

  In this case, it is possible to cope with not only the change of the transfer characteristic due to the occupant position but also the change of the transfer characteristic due to the temperature. At this time, similarly to the detection of the occupant position, by using any one of the piezo actuators 5a to 5d arranged at a position where the transfer characteristics from the piezo actuator 5 to the acceleration sensor 4 are likely to vary depending on the temperature, It becomes easy to distinguish from the transfer characteristics, and it becomes easier to specify the temperature of the vehicle body or the temperature in the passenger compartment with higher accuracy.

  Next, a second embodiment of the present invention will be described. In Example 1, the test signal was applied to one piezoelectric actuator selected from the piezoelectric actuators 5a to 5d. In the second embodiment, test characteristics are applied to a plurality of piezo actuators to detect transfer characteristics. In addition, about the part similar to Example 1, the detailed description is abbreviate | omitted by attaching | subjecting the same code | symbol.

  The test signal generator 334 applies a test signal to a plurality of piezo actuators selected from the piezo actuators 5a to 5d, and sets a signal applied to the other piezo actuators to zero. A test signal applied to each piezoelectric actuator uses a sine wave signal having a single frequency. By applying test signals to a plurality of piezo actuators in this way, the state estimation unit 332 compares the transfer characteristics, and when identifying an occupant position having a close transfer characteristic, the number of transfer characteristics to be compared increases. The occupant position can be detected more accurately.

  Here, the frequency of the test signal may be different among a plurality of piezoelectric actuators, or may be the same. For example, the test signal may be set as follows. The frequency of the test signal is set to a frequency having a low transfer characteristic from each piezo actuator to the control space sound pressure, regardless of the occupant positions A, B, and C showing the transfer characteristic as shown in FIG. By applying a test signal having such a frequency, the level of sound generated by the test signal is lowered, and noise due to the test signal can be reduced.

  The test signal may be applied to the piezo actuator 5 disposed at a position where the transfer characteristic from the piezo actuator 5 to the acceleration sensor 4 is likely to vary depending on the occupant position. By applying a test signal to such a combination of piezo actuators 5, when comparing the transfer characteristics of the piezo actuators 5, differences from other occupant position states can be easily detected. The occupant position can be specified with high accuracy. Such a piezo actuator 5 is, for example, installed under the seat or behind the seat. These piezo actuators can easily detect the occupant of the seat because the transmission characteristics differ greatly depending on whether or not the occupant of the seat is installed.

  When a test signal is applied to a plurality of piezo actuators 5, vibrations caused by the test signal are detected by the acceleration sensor 4, but the test signal that cancels noise caused by the test signal in the control space 6 is as follows. To generate.

  FIG. 10 is a block diagram for calculating a filter that generates a test signal that cancels out noise caused by the test signal in the control space 6. Here, one of the test signals applied to the plurality of piezo actuators 5 is set as a reference signal ui. When the frequency band of the test signal is different for each piezo actuator 5, the reference signal may be changed for each frequency and the corresponding test signal may be used as the reference signal.

  In FIG. 10, a block 3331 (denoted as Ct in FIG. 10) indicates a filter that generates a signal that cancels noise caused by the reference signal. A block 3332 (denoted as Gpj-SPL in FIG. 10) represents a transfer function from the piezo actuator 5 that outputs a wave that cancels noise caused by the reference signal to the control space sound pressure. A block 3333 (described as Gpi-SPL in FIG. 10) represents a transfer function from the piezo actuator 5 that outputs a test signal that is a reference signal to the control space sound pressure. A block 3334 (denoted as W1 in FIG. 6) represents a weight function for the output of the piezoelectric actuator 5.

  The transfer functions of the block 3332 and the block 3333 may be system-identified by the same method as at the time of designing the control command signal generation filter described in the first embodiment. However, the transfer functions of the block 3332 and the block 3333 need to be changed depending on the occupant position, and may be set according to the occupant position estimated by the state estimation unit 332 at the previous time. In addition, after estimating a passenger | crew position in the state estimation part 332, the block 3332 and the block 3333 are changed according to the result.

  Block 3334 sets the control command signal so that it does not exceed the maximum voltage of the piezoelectric actuator 5. At this time, the filter Ct 3331 is designed so as to minimize the transfer function from the reference signal ui to the control space sound pressure level (SPL) and the norm of the transfer function from the reference signal ui to the reference signal cancellation signal uj. The design of the filter Ct 3331 may be performed in the same manner as the control command signal generation filter 335 described in the first embodiment.

  Next, the process of the control command signal calculation unit 33 will be described with reference to FIG. In step S <b> 201, the transfer characteristic identification unit 331 checks whether the test signal is applied by the test signal generation unit 334. When the test signal is applied, the piezoelectric actuator to which the test signal is applied and its frequency are specified, and the process proceeds to S202. When the test signal is not applied, the characteristic changing unit 333 selects the same control command signal generation filter 335 as the previous one, and the process ends.

  In step S <b> 202, the transfer characteristic identification unit 331 identifies the transfer characteristic from the test signal applied to each piezoelectric actuator 5 to the acceleration detection signal. Actually, the input calculated by the control command signal generation filter 335 for controlling the noise is also applied to each piezo actuator 5, so that the adder 336 outputs the test signal and the output of the control command signal generation filter 335. The characteristics from the added control command signal to the acceleration detection signal are identified. However, since the transmission path is the same, the identification result of the transmission characteristic coincides with the case where only the test signal is applied.

  For identification of this transfer characteristic, a non-parametric system identification method such as “weighted overlapped segment averaging method” may be used. The system identification may be limited to only the piezoelectric actuator to which the test signal specified by the transfer characteristic identification unit 331 is applied and the frequency to which the test signal is applied. By limiting the frequency to be identified, the time required for the calculation can be reduced.

  In step S203, the state estimation unit 332 compares the transfer characteristic for each occupant position combination stored in advance in the memory in the state estimation unit 332 with the result of system identification in step S202. The transmission characteristics for each occupant position combination stored in the memory are determined by applying an impulse signal or white noise to the piezo actuator 5 for each possible occupant position, and using the acceleration signal obtained at that time to perform system identification. It can be obtained by doing. For system identification, a method similar to that used when applying a test signal may be used.

  In step S204, the state estimation unit 332 determines whether there is a transfer characteristic close to the result of system identification by the transfer characteristic identification unit 331 and the transfer characteristic for each combination of passenger positions stored in advance. To do. If there is an occupant position combination with similar transfer characteristics, the process moves to step S205. If there is no occupant position combination characteristic with a similar transfer characteristic, the process moves to step S206. Here, for example, the determination as to whether there is an occupant position combination with similar transfer characteristics may be performed as follows. The difference between the transfer characteristic at the current time identified in step S202 and the transfer characteristic at each occupant position combination is calculated for each piezo actuator to which the test signal is applied. Then, the sum of absolute values of the differences in the transfer characteristics is calculated, and the passenger position combination having the smallest value is set as the passenger position at the current time. However, it is determined that there is no close occupant position combination when either the gain or the phase of the difference in the transfer characteristics of the piezoelectric actuators exceeds the threshold value Tg and the threshold value Tp, respectively. When performing noise control, the threshold Tg and the threshold Tp may be selected so that a desired control effect can be obtained even if the characteristics deviate by the threshold Tg and the threshold Tp, respectively, from the transfer characteristics at the time of filter design.

  In step S205, the characteristic changing unit 335 switches the control command signal generation filter 335 corresponding to the estimated occupant position combination from the filters stored in the memory in advance, and ends the process.

  In step S206, it is determined that there is a factor that causes the transfer characteristic to fluctuate other than the difference in the occupant position, and the process ends without switching the filter.

  Here, the method for detecting the occupant position has been described in the present embodiment, but not only the transfer characteristics for each occupant position combination but also the transfer characteristics for temperature differences are stored in the memory in the state estimation unit 332, The closest transfer characteristic may be selected in comparison with the above characteristic. Then, the control command signal generation filter 335 corresponding to the occupant position and temperature may be selected by the characteristic changing unit 333.

  In this case, it is possible to cope with not only the change of the transfer characteristic due to the occupant position but also the change of the transfer characteristic due to the temperature. At this time, similarly to the detection of the occupant position, by using a piezo actuator in which the transfer characteristic from the piezo actuator 5 to the acceleration sensor 4 is likely to vary depending on the temperature, it becomes easier to distinguish the transfer characteristic from different temperatures. It becomes easy to specify the temperature of the vehicle body and the passenger compartment with high accuracy.

  Here, the first embodiment and the second embodiment may be performed simultaneously to detect the occupant position and temperature, and the control command signal generation filter 335 may be changed based on the result. In this case, it is possible to specify the occupant position and the temperature with higher accuracy than when the first embodiment and the second embodiment are individually performed.

  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.

  According to the noise reduction devices of the first and second embodiments described above, the following operational effects can be obtained.

  Without providing sensors to detect the state of the structure, the characteristics of the drive signal of the wave application unit that applies the wave to the structure can be changed according to the state change of the structure, making the device complicated Without lowering the cost and without increasing the cost, it is possible to prevent a reduction in noise reduction performance due to a change in the state of the structure.

  In addition, when the occupant position, the temperature of the vehicle body and the passenger compartment change, and the transfer characteristics of the control space from the wave application section change, control means that suppresses the deterioration of the noise control effect due to the change in the characteristics is used. The uncomfortable feeling felt by the occupant can be reduced by the deterioration of the control effect.

  In addition, since the test signal is applied to the wave application unit in which the acceleration sensor output characteristic is likely to change due to the change in the position of the person in the structure, the position of the passenger can be easily determined from the transfer characteristic. For this reason, a control means can be changed more appropriately, the deterioration of the noise control effect by a characteristic change can be suppressed, and the discomfort which a passenger | crew feels can be reduced.

  In addition, since the test signal is applied to the wave application unit located on the lower or rear surface of the seat installed in the structure, there is a large difference in transfer characteristics depending on the passenger position. It becomes easy to discriminate and the control means can be changed more appropriately.

  In addition, because the test signal is applied to the structure and the wave application unit where the acceleration sensor output characteristics are likely to change due to temperature changes in the structure, there is a large difference in transmission characteristics due to changes in the body temperature and the cabin temperature. It becomes easier to determine the occupant position from the transfer characteristics.

  In addition, since the acceleration sensor output characteristics for each personnel position are stored in the memory in advance and the personnel position is estimated by comparing with the acceleration sensor output characteristics based on the test signal, the transfer characteristics stored in the memory are This makes it easier to determine the occupant position from the current transmission characteristics.

  Further, since the control space is changed based on the personnel position estimated from the acceleration sensor output characteristics, the noise control can be concentrated in the space where the passenger is present, and the noise control can be performed more effectively.

  Also, the acceleration sensor output characteristics for each temperature in the structure and the structure are stored in advance, and by comparing with the acceleration sensor output characteristics by the test signal, it becomes easier to discriminate the vehicle body temperature and the cabin temperature. The deterioration of the noise control effect due to the state change can be suppressed, and the discomfort felt by the occupant can be reduced.

  Further, since the test signal includes a plurality of frequency components, the frequency band to be compared increases when the occupant position is specified, so that the occupant position can be detected more accurately. As a result, the control means can be changed more appropriately, the deterioration of the noise control effect due to the characteristic change can be suppressed, and the discomfort felt by the occupant can be reduced.

  Further, by applying the test signal to a plurality of piezo actuators (wave application units), the transmission characteristics to be compared are increased when specifying the occupant position, so that the occupant position can be detected more accurately. As a result, the control unit can be changed more appropriately, the deterioration of the noise control effect due to the state change of the structure can be suppressed, and the discomfort felt by the occupant can be reduced.

  In addition, the test signal is a frequency signal having a low transfer characteristic gain from the wave application unit to the control space sound pressure regardless of the position and temperature of the person, so that when the test signal is applied and the transfer characteristic is identified, The test signal is less likely to be noise, and the discomfort felt by the occupant can be reduced.

  In addition, by applying a test signal outside the audible band at the time of key-on, the test signal does not become noise when the transfer characteristic is identified, and the discomfort felt by the occupant can be reduced.

  Next, a third embodiment of the noise reduction device according to the present invention will be described. The overall configuration of the third embodiment is the same as that of the first and second embodiments shown in FIG. In addition, the same code | symbol is provided to the part which is common in Example 1, 2, and the overlapping description is abbreviate | omitted.

  FIG. 12 shows a block diagram of a noise control system including the control unit 30 in the third embodiment of the noise reduction apparatus according to the present invention. In FIG. 12, d is vibration at the position of the acceleration sensor 4 caused by the tire excitation force, α is an acceleration detection signal of the acceleration sensor 4 that is an input of the control unit 30, and u is the control unit 30 to the piezo actuator 5. The control command signal to be output, SPL is the sound pressure in the control space, C (s) is the transfer function of the control unit 30, and Gd-SPL (s) is the transfer from the vibration d due to the tire excitation force to the control space sound pressure SPL. The function, Gu-α (s), is a transfer function from the control command signal u to the acceleration detection signal α, and Gu-SPL (s) is a transfer function from the control command signal u to the control space sound pressure SPL.

  As shown in FIG. 12, the vibration of the piezo actuator 5 driven by the control command signal u propagates through the floor panel 3 with the characteristic represented by the transfer function Gu-α (s), and the vibration from the tire 2. d is added and detected as an acceleration detection signal α and input to the control unit 30. Further, the vibration of the piezo actuator 5 driven by the control command signal u propagates through the floor panel and the vehicle interior space with the characteristic represented by the transfer function Gu-SPL (s), and becomes a control sound in the control space 6. . This control sound acts on the road noise noise in the control space 6 expressed by the transfer function Gd-SPL (s) from the vibration d at the sensor position caused by the tire excitation force, and this noise is reduced. .

  The transfer function Gu-SPL (s) from the control command signal u, which is an input of the piezo actuator 5, to the control space sound pressure SPL changes according to the temperature of the vehicle body or the interior space. Therefore, if the temperature assumed at the time of designing the control unit 30 is different from the actual temperature, the transfer function Gu-SPL (s) changes, and the control sound in the control space 6 deviates from the desired control sound. May cause an error, reducing the noise reduction effect in the control space 6 and giving the passengers in the vicinity of the control space 6 an uncomfortable feeling. In addition, since the change in the transfer function depends on the temperature of the vehicle body and the interior space, in order to measure the temperature change that causes the change of the transfer function, it is necessary to install a plurality of temperature sensors in the vehicle body and the interior space. To rise. Further, a method of measuring the input / output signal of the transfer function Gu-SPL (s) and identifying the change of the transfer function Gu-SPL (s) from the relationship between these signals can be considered. In this embodiment, the control space 6 This method cannot be used because no microphone is installed.

  Here, as shown in FIG. 3B, the transfer function Gu-α (s) from the control command signal u, which is the input of the piezo actuator 5 (wave application unit), to the acceleration detection signal α of the acceleration sensor 4 is also obtained. It changes according to the temperature of the vehicle body, and it is known that there is a relationship between the temperature changes of the transfer function Gu-SPL (s) and the transfer function Gu-α (s).

  Therefore, in this embodiment, the temperature change of the transfer function Gu-α (s) is identified from the control command signal u and the acceleration detection signal α, and the change of the transfer function GGu-α (s) is used using this relationship. From this, the temperature change of the transfer function Gu-SPL (s) is estimated, and the control unit 30 is adjusted according to this change, thereby suppressing the deterioration of the control effect due to the temperature change without using a microphone or a temperature sensor. It is possible to reduce the discomfort of the passengers near the control space 6.

  FIG. 13 is a diagram that defines the number of acceleration sensors 4, piezoelectric actuators 5, and control spaces 6 that are assumed in the third embodiment. As shown in FIG. 13, the acceleration sensor 4 is premised on four of 4a, 4b, 4c and 4d, the piezoelectric actuator 5 is premised on three of 5a, 5b and 5c, and the control space is premised on two of 6a and 6b. . As described above, when there are a plurality of tire excitation forces, a plurality of acceleration sensors, a plurality of piezoelectric actuators, and a plurality of control spaces, each acceleration sensor has a vibration caused by all the tire excitation forces and all the piezoelectric actuator waves. Are detected in an overlapping manner, and in each of the control spaces 6a and 6b, all the tire excitation forces and all the sounds generated by the piezo actuator waves are generated in an overlapping manner.

  Note that the number of acceleration sensors, piezoelectric actuators, and control spaces is not limited to the number shown in FIG. The number of acceleration sensors 4 is generally required to be larger than the number of tires 2 that are vibration sources. The number and installation positions of the specific acceleration sensors 4 on the floor panel 3 are determined so that the coherency between each acceleration sensor 4 and the sound pressure of noise in the control space 6 is sufficiently high (for example, 0.9 or more). do it.

  A sufficient number of piezoelectric actuators 5 are attached to appropriate positions on the floor panel 3 of the vehicle body in order to reduce noise in the control space 6.

  For the control space 6, for example, several control space candidates are set in advance, and a candidate close to the occupant position is appropriately selected as a control space using a seat pressure sensor or an infrared sensor that detects the occupant position in the passenger compartment. do it. Further, a predetermined space may be determined as a control space in advance without using a sensor.

  In addition to the above, since all noises generated from the floor panel 3 are included as control targets, part of the engine noise and wind noise generated by the air flowing through the bottom of the vehicle body can be handled in the same manner.

  Further, 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 3, but for example, noise in the passenger compartment generated by the same mechanism such as a dash panel, a windshield, and a roof panel (both not shown). The same effect can be obtained for the generation source if the present invention is used for the site.

  The overall schematic diagram of the noise reduction apparatus according to this embodiment is the same as that of the first embodiment shown in FIG. FIG. 14 shows a control block diagram of this embodiment of the control command signal calculation unit 33 in FIG.

  The control command signal calculation unit 33 of the present embodiment is based on the noise control unit 41 that reduces the predetermined spatial sound pressure based on the acceleration detection signal and the test signal when the test signal is input to the piezo actuator 5 (wave application unit). A vibration characteristic change estimation unit 42 that estimates or detects a change in vibration transfer characteristic up to the acceleration detection signal, and a sound from the piezoelectric actuator 5 input to a predetermined spatial sound pressure based on the estimated or detected change in vibration transfer characteristic. A sound vibration characteristic change estimation unit 43 that estimates a vibration characteristic change, a characteristic change compensation unit 44 that changes a characteristic of the noise control unit 41 according to the sound vibration characteristic change, an output of the noise control unit 41, and a test signal And an adder 45 for adding

  In this embodiment, the control command signal calculation unit 33 is mounted on a so-called digital computer, and the calculation is executed, for example, every control cycle of 1 msec. This digital computer includes, for example, a CPU, a program ROM, a working RAM, and an input / output interface.

  The noise control unit 41 uses the transmission characteristic C of the control command signal calculation unit 33 from the acceleration detection signal α detected by the acceleration sensor 4 to reduce the noise control command to the piezo actuator 5 so as to reduce the noise in the control space 6. The value uc is calculated. The relationship between the acceleration detection signal α and the noise control command value uc is expressed by the following equations (2) and (3).

x = Ac × x + Bc × α (2)
uc = Cc × x + Dc × α (3)
Here, Ac, Bc, Cc and Dc are parameters representing the transfer characteristic C of the control command signal calculation unit 33, and the noise control command value uc is obtained from the acceleration detection signal α using the equations (2) and (3). Calculated.

  The vibration characteristic change estimation unit 42 sets a test signal us to be added to the noise control command value uc to the actuator, and based on the test signal us and the acceleration detection signal α, a transfer function from the test signal us to the acceleration detection signal α. Is identified. The relationship among the noise control command value uc, the test signal us, and the actuator input signal u is expressed by the following equation (4).

u = uc + us (4)
Therefore, from this linear relationship, the transfer function from the test signal us to the acceleration detection signal α is equal to the transfer function Gu−α (s) from the actuator input signal u to the acceleration detection signal α.

  Therefore, for example, white noise is input as a test signal to any one of the piezoelectric actuators 5a to 5c, and the piezoelectric signal to which the test signal is applied from the white noise and the output of any one of the acceleration sensors 4a to 4d. The transfer function Gu-α (s) from the actuator input to the selected acceleration sensor output is obtained. Next, the obtained transfer function Gu-α (s) is compared with the transfer function for each temperature stored in advance in the memory, and the temperature of the transfer function most suitable is obtained. Next, a temperature difference between the precondition temperature when the noise control unit 41 is designed and the obtained temperature is output to the sound vibration characteristic change estimation unit 43 as a feature amount.

  Further, in order to determine the necessity of changing the noise control unit 41 by the characteristic change compensation unit 44, the transfer functions Gu-α (s) from all the piezo actuators 5 to the acceleration sensors 4 are measured and sent to the characteristic change compensation unit 44. Output.

  Here, the test signal is not limited to white noise, and any test signal that can estimate the change of the transfer function Gu-α (s) from the piezo actuator input to the acceleration sensor output due to a temperature change or the like may be used. For example, a sine wave signal having a single frequency or a certain bandwidth may be input, and this frequency may be gradually changed to measure the transfer function Gu-α (s) in the entire frequency band of interest.

  Further, such a sine wave signal having a single frequency or a certain bandwidth may be output not limited to the entire frequency band but limited to a frequency sufficient to estimate the change of the transfer function Gu-α (s).

  For example, when the test signal is output only at a frequency at which the sound vibration characteristic gain is near the anti-resonance point and the temperature change can be read from the vibration characteristic gain as in the frequency ω shown in FIG. Since the change of the transfer function Gu-α (s) at the frequency ω can be estimated while reducing the noise due to the test signal, the passenger's discomfort due to the test signal noise can be reduced.

  In addition, if the test signal is output at a frequency ω in the vicinity of the anti-resonance frequency of the acceleration sensor detection signal frequency characteristic before the test signal is output, the S / N ratio of the test signal can be increased, so that the transfer function Gu-α (s) For example, the accuracy of compensation for noise changes due to temperature changes is increased, and passenger discomfort can be further reduced.

  Then, for example, a temperature change may be read from the change of the transfer function Gu-α (s) at the frequency ω and output as a feature amount. For example, the frequency change of the transfer function Gu-α (s) corresponding to the current temperature is selected from the frequency characteristics of the transfer function Gu-α (s) for each temperature stored in the memory in advance. It may be output to the compensation unit 44.

Note that the feature amount is not limited to a temperature change, and may be another factor that changes the transfer function Gu-α (s) and the transfer function Gu-SPL with relevance.
Further, the noise control signal uc may be considered as the test signal without using the test signal.

  Further, when the test signal is generated at the time of opening a window or at the time of audio output, noise in the control space by the test signal can be masked, and passenger discomfort can be reduced. Furthermore, it is still effective to set the test signal level below the audio volume. Further, when the test signal is output when the vehicle is stopped, no input from the road surface is input, so the S / N ratio of the test signal is improved and the identification accuracy of the transfer characteristic is improved.

  The sound vibration characteristic change estimation unit 43 outputs the transfer function Gu-SPL stored in advance in the memory to the characteristic change compensation unit 44 according to the feature amount input from the vibration characteristic change estimation unit 42. For example, when the feature amount is temperature, the transfer function Gu-SPL prepared for each temperature may be selected and output according to the temperature.

  In the characteristic change compensator 44, the transfer function Gu-α (s) input from the vibration characteristic change estimator 42, the transfer function Gu-SPL input from the sound vibration characteristic change estimator 43, and the noise used at the current time. FIG. 12 shows the transfer function C (s) of the control unit 41, the transfer function Gd-SPL from the disturbance d to the control space sound pressure acquired in advance and stored in the memory, and the acceleration detection signal. In order to increase the noise reduction effect, for example, if the control performance change at the time of the characteristic change between the transfer function Gu-α (s) and the transfer function Gu-SPL is obtained, The noise control unit 41 is changed. Here, the predetermined value may be, for example, 3 dB which is said to be the quietness on the second class of the sedan passenger car, but is not limited to this value.

  The relationship between the disturbance d (vibration caused by the tire excitation force), the control command signal u, and the acceleration detection signal α input to the block diagram is expressed by the following equation (5).

α = d + Gu−α × u (5)
Therefore, the disturbance d inputted to the block diagram is obtained from the control command signal u and the acceleration detection signal α using this equation (5).

  The control unit 30 is redesigned using, for example, the following H2 control. First, the transfer function C (s) and the transfer function Gu-α (s) are extracted from the noise control system shown in FIG. 12 and deformed as shown in FIG. Here, Wc (s) is a weight function of the piezoelectric actuator output, u is a control command signal, and y is an acceleration detection signal. The reason for extracting Gu-α (s) is to design the transfer function C0 (s) on the assumption that there is no vibration wraparound from the control command signal u to the acceleration detection signal α.

  When designing with H2 control using the generalized plant shown in FIG. 16, the transfer function C (s) is designed so as to minimize the H2 norm from disturbance d (vibration caused by tire excitation force) to z1 and z2. Therefore, a transfer function C0 (s) that reduces road noise is obtained. For H2 control, see Active Structural Acoustic Control in a car cabin using a virtual sound sensing method, Mens Remy Shell (Nissan Motor), Yoshiro Takamatsu, Satoshi Deguchi, Haruki Yashiro, Dynamics and Design Conference 2006 (D & D2006).

  Actually, there is a vibration wraparound from the control command signal u, which is a piezo actuator command value, to the acceleration detection signal α. Therefore, in order to eliminate this wraparound, as shown in FIG. By using the estimation model eGu-α (s) of the transfer function Gu-α (s), the amount of sneak from the control command signal u is estimated and subtracted from the acceleration detection signal α. A disturbance d, which is an input of the transfer function C0 (s), is extracted. Therefore, as shown in FIG. 17, the combination of the transfer function C0 (s) of the noise control unit 41 and the estimated model eGu-α is set as the transfer function C (s) of the noise control unit 41. Thereby, the deterioration of the control effect due to the change of the transfer function Gu-α (s) can also be suppressed.

  Note that the method of changing the noise control unit 41 is not limited to this method, and the above design is performed for each feature quantity such as temperature, for example, and stored in the memory in advance. The control unit 41 may be selected.

  When the noise control unit 41 is designed as described above, the noise control unit 41 is changed in consideration of the frequency characteristics of the transfer function Gu-α (s) and the transfer function Gu-SPL. It can compensate not only for the rise, but also for sound quality changes.

  Next, the calculation executed by the control command signal calculation unit 33 will be described with reference to the flowchart of FIG. In step S301, the vibration characteristic identification calculation described in the description of the vibration characteristic change estimation unit 42 is performed, and the process proceeds to step S302.

  In step S302, the calculation described in the description of the sound vibration characteristic change estimation unit 43 is performed, and the process proceeds to step S303.

  In step S303, as shown in the description of the characteristic change compensation unit 44, for example, a change in noise reduction effect due to a temperature change is estimated by simulation, and the process proceeds to step S304.

  In step S304, as shown in the description of the characteristic change compensation unit 44, if the noise reduction effect estimated in step S303 is equal to or less than a predetermined value, the process proceeds to step S305 to change the noise control unit 41 to increase the noise reduction effect. If the estimated noise reduction effect is not less than the predetermined value, the process proceeds to step S306 to maintain the current state.

  In step S305, as shown in the description of the characteristic change compensation unit 44, for example, the noise control unit 41 is changed to compensate for a decrease in noise reduction effect due to a temperature change, and the process proceeds to step S306.

  In step S306, the noise control unit 41 calculates the control command signal u based on the acceleration detection signal α and ends the process.

  According to the third embodiment described above, the temperature change characteristics of the sound vibration characteristic and the vibration transfer characteristic from the wave applying unit (piezo actuator) to the predetermined spatial sound pressure are as shown in FIG. Using this relationship, the vibration transfer characteristics change to the acceleration sensor output when a test signal is input to the wave application unit is measured, and the sound vibration from the wave application unit input to the specified spatial sound pressure is measured from this change. By estimating the characteristic change, for example, it is possible to estimate a sound vibration characteristic change including a change in solid propagation of the vehicle body due to a temperature change. And since the noise control unit is changed according to this sound vibration characteristic change, it is possible to suppress a decrease in noise reduction effect due to a temperature change or the like without providing a temperature sensor or the like, and to reduce passenger discomfort due to noise deterioration. effective.

  Further, according to the third embodiment, since the test signal is generated in the vicinity of the anti-resonance frequency of the sound vibration characteristic of the structure, there is an effect that occupant discomfort due to the test signal noise can be reduced.

  Further, according to the third embodiment, the test signal is generated in the vicinity of the anti-resonance frequency of the acceleration sensor output signal frequency characteristic before the test signal is input, so that the S / N ratio of the acceleration detection signal when the test signal is input is increased, The estimation accuracy of the transfer function Gu-α (s) is increased, and for example, there is an effect that the compensation accuracy of noise change due to temperature change can be improved, and passenger discomfort can be further reduced.

DESCRIPTION OF SYMBOLS 1 Car body 2 Tire 3 Floor panel 4 Acceleration sensor (vibration detection means)
5 Piezo actuators (wave application means)
6 control space 30 control unit 31 amplifier 32 A / D conversion unit 33 control command signal calculation unit 331 transfer characteristic identification unit 332 state estimation unit 333 characteristic change unit 334 test signal generation unit 335 control command signal generation filter 336 addition unit 34 D / A conversion unit 35 amplification unit 41 noise control unit 42 vibration characteristic change estimation unit 43 sound vibration characteristic change estimation unit 44 characteristic change compensation unit 45 addition unit

Claims (17)

Vibration detecting means for detecting the vibration of the structure;
Wave applying means for applying a wave to the structure;
By generating a drive signal to be supplied to the wave applying means according to the detection signal of the vibration detecting means, and applying a control wave to the structure from the wave applying means, at least vibration and noise of the structure Control means to reduce one,
In a noise reduction device equipped with
The control means includes
A transfer characteristic identifying means for identifying a transfer characteristic from the test signal to the detection signal when a test signal is input to the wave applying means;
State estimating means for estimating the state of the structure according to the transfer characteristics;
Characteristic changing means for changing the characteristic of the drive signal in accordance with the estimated state;
A noise reduction device comprising:   The noise reduction device according to claim 1, wherein the state of the structure is a staff position in the structure.   The noise reduction device according to claim 1, wherein the state of the structure is a temperature in the structure.   4. The test signal according to claim 1, wherein the test signal is supplied to a wave applying unit disposed at a position where the output characteristic of the detection signal is likely to change due to a change in personnel arrangement in the structure. The noise reduction device according to item 1. The wave applying means for supplying the test signal includes:
5. The noise reduction device according to claim 4, wherein the noise reduction device is a wave applying unit disposed on a lower surface or a rear surface of a seat installed in the structure.   4. The test signal is supplied to a wave applying unit disposed at a position where an output characteristic of the detection signal is easily changed by a temperature change in the structure and the structure. The noise reduction device according to any one of the above. The state estimating means includes
Comparing the output characteristics of the detection signal for each personnel position stored in advance with the output characteristics of the detection signal when the test signal is supplied to the wave applying means, and estimating the personnel position closest to the output characteristics. The noise reduction device according to claim 4 or 5, characterized by the above.   8. The control unit according to claim 7, wherein the control unit changes a control space, which is a target space for reducing at least one of vibration and noise of the structure, based on the personnel position estimated by the state estimation unit. Noise reduction equipment.   Comparing the output characteristics of the detection signal for each temperature in the structure and the structure stored in advance with the output characteristics of the detection signal based on the test signal, and estimating the temperature closest to the output characteristics The noise reduction device according to claim 3 or 6.   The noise according to any one of claims 1 to 9, wherein the test signal includes a signal obtained by multiplexing a plurality of frequency components or a signal in which a plurality of frequency components are arranged in time series. Reduction device.   The noise reduction device according to any one of claims 1 to 10, wherein the test signal is supplied to a plurality of wave applying means.   12. The test signal according to claim 1, wherein the test signal is a signal having a frequency with a low gain of a transfer characteristic from the wave applying means to the detection signal regardless of a staff position and temperature. The noise reduction device described in 1.   2. A noise reduction device mounted on an internal combustion engine vehicle, wherein a test signal outside an audible band is supplied to the wave applying means from when a key is inserted to when the engine is started. The noise reduction device according to any one of 12. Vibration detecting means for detecting the vibration of the structure;
Wave applying means for applying a wave to the structure;
By generating a drive signal to be supplied to the wave applying means according to the detection signal of the vibration detecting means, and applying a control wave from the wave applying means to the structure, at least vibration and noise of the structure Noise control means to reduce one,
In a noise reduction device equipped with
Vibration characteristic change estimating means for estimating or detecting a change in vibration transfer characteristics from the test signal to the detection signal when a test signal is input to the wave applying means;
Sound vibration characteristic change estimating means for estimating a sound vibration characteristic change from the wave applying means input to a predetermined spatial sound pressure based on the estimated or detected change of the vibration transfer characteristic;
A characteristic change compensating means for changing a characteristic of the noise reducing means in accordance with the sound vibration characteristic change;
A noise reduction device comprising:   The noise reduction device according to claim 14, wherein the test signal is a signal having an anti-resonance frequency of a structure.   15. The noise reduction device according to claim 14, wherein the test signal is generated in the vicinity of an anti-resonance frequency of an output signal frequency characteristic of the vibration detection means before the test signal is input. In a noise reduction method for reducing at least one of vibration and noise of the structure by applying a wave to the structure according to a detection signal that detects vibration of the structure,
Based on the detection signal when a test wave is applied to the structure, the transfer characteristic from the test wave to the detection signal is identified,
Estimating the state of the structure according to the transfer characteristics;
A noise reduction method comprising changing a characteristic of a wave applied to the structure according to the estimated state.

Images