Patent Translate Powered by EPO and Google Notice This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate, complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or financial decisions, should not be based on machine-translation output. DESCRIPTION JPH07162995 [0001] BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a hydraulic drive speaker, and more particularly to a hydraulic drive speaker with improved frequency characteristics. [0002] 2. Description of the Related Art An apparatus for converting an electric signal to voice includes a speaker, and the speaker includes an electrodynamic actuator type speaker and a hydraulic pressure type speaker. [0003] As is well known, a dynamic actuator type speaker vibrates a cone by driving a voice coil by an electric signal and converts it into sound. [0004] On the other hand, a hydraulic drive speaker controls the servo valve with an electric signal to change the pressure of liquid in the cylinder, thereby vibrating the diaphragm attached to the piston head and converting it into sound. . [0005] A hydraulic drive speaker is devised as a loudspeaker for reproducing a deep bass, and has the 05-05-2019 1 following structural features in comparison with a conventional speaker (electrodynamic actuator method). [0006] (1) Since the diaphragm has an edgeless structure and the diaphragm is driven by the hydraulic piston, the stroke can be increased. Therefore, low frequency sound can be reproduced. [0007] (2) Since a large input can be made using the hydraulic driving force, a large sound pressure output can be obtained. [0008] However, the hydraulic drive type speaker has a problem that although it can emit an extremely low frequency sound of 100 Hz or less, it is difficult to emit a further sound. [0009] Therefore, an object of the present invention is to solve the above-mentioned problems, and to provide a hydraulic drive type speaker capable of producing a sound of a frequency from extremely low frequency to several hundreds Hz. [0010] SUMMARY OF THE INVENTION In order to achieve the above object, according to the present invention, a diaphragm, a piston rod attached to the diaphragm, and a liquid in which the piston rod is inserted and reciprocably supported A pressure cylinder, a piston head integrally formed on the outer periphery of the piston rod to divide the inside of the cylinder into two front and rear chambers, controlling the amount of oil supplied to each cylinder chamber according to a control input signal and A servo valve that vibrates the diaphragm, a displacement detector that detects displacement of the diaphragm and applies position feedback so that the diaphragm maintains the home position, and a diaphragm that is an output signal of the control target An acceleration detector for detecting an acceleration, and a compensation controller for receiving 05-05-2019 2 an audio input signal and outputting a control input signal to the servo valve based on the audio input signal; After the compensation controller actually measures the inherent frequency characteristics generated by the diaphragm driven by the servo valve, hydraulic cylinder, and piston rod for the voice input signal that has been known in advance by the acceleration detector, the target frequency range In order to flatten the output frequency characteristics at the same time, based on H 補償 theory, it is configured to perform compensation control for compensating for the inherent frequency characteristics, and the voice input signal and acceleration actually input by the compensation controller The control input signal is generated based on the signal to drive the servo valve. [0011] Further, the compensation controller of the hydraulic drive speaker according to the present invention is given by equation 1 in order to flatten the frequency characteristic of the complementary sensitivity function V (s) in the low frequency range. [0012] (Where the complementary sensitivity function V (s) is (1 + GK) -1, and G is a model after measuring the natural frequency characteristics of the input signal with respect to the diaphragm) The transfer frequency function of the object to be controlled, K is a rational function derived from G on the basis of H∞ theory, and WS is a weighting function of the sensitivity function V) has a compensation frequency characteristic which minimizes the function. [0013] Further, the compensation controller of the hydraulic drive speaker according to the present invention can use the equation (2) to flatten the frequency characteristic of the complementary sensitivity function V (s) in the low frequency range. [0014] (Where the complementary sensitivity function V (s) is (1 + GK) -1 GK, and G is a hydraulic speaker body consisting of a piston rod, a hydraulic cylinder, a piston head and a servo valve) Transfer function of hydraulic speaker body obtained by modeling after measuring frequency characteristics, K is a rational function derived from G∞H∞ theory, WT is a weight function of complementary sensitivity V), Have compensation frequency characteristics. [0015] Further, the compensation controller of the hydraulic drive speaker according to the present 05-05-2019 3 invention is the equation 3 in order to flatten the frequency characteristic of the complementary sensitivity function V (s) in the low frequency range. [0017] (Where the complementary sensitivity function T and the sensitivity function S are [0019] G is a transfer function of the fluid pressure speaker body obtained by measuring and modeling the frequency characteristics of the fluid pressure speaker body consisting of a piston rod, a fluid pressure cylinder, a piston head and a servo valve, and K is from G to H有 A rational function derived based on the theory, WS and WT respectively have a compensation frequency characteristic that minimizes the sensitivity function S and the weighting function of the complementary sensitivity function T). [0020] According to the above configuration, the compensation controller designed on the basis of the H∞ theory is equivalent to the sound of the sound output from the speaker main body consisting of the servo valve, the hydraulic cylinder, the piston rod and the diaphragm. Since the control input signal is generated based on this deviation and the derived compensation frequency characteristic to drive the servo valve, the deviation of the voice input inputted to the compensation controller is detected. The frequency characteristic is corrected by the compensation controller having the opposite frequency characteristic and then input to the hydraulic speaker body to shape the frequency characteristic into a flat. The diaphragm acceleration corresponding to the sound output generated from the diaphragm based on the shaped frequency characteristic is detected by the acceleration detector and input to the compensation controller, and the fluid pressure speaker body and the compensation controller based on the deviation thereof And the transfer function of the compensation controller is controlled so that the frequency characteristic of the closed loop formed by Further, since position feedback is applied to the diaphragm, it is possible to prevent the position of the diaphragm from being biased to one side. 05-05-2019 4 [0021] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will now be described in detail with reference to the attached drawings. [0022] FIG. 1 (a) is a conceptual view of one embodiment of a hydraulic drive speaker according to the present invention. [0023] In FIG. 1 (a), 1 is a diaphragm and 2 is a piston rod attached to the diaphragm 1 for generating vibration. Reference numeral 3 denotes a hydraulic cylinder which inserts the piston rod 2 and supports it in a reciprocating manner, and 4 denotes a piston which is integrally formed on the outer peripheral surface of the piston rod 2 and divides the inside of the hydraulic cylinder 3 into two front and rear chambers. It is a head. A servo valve 5 controls the amount of liquid supplied to the cylinder chambers 3a and 3b. The piston rod 2, the hydraulic cylinder 3, the piston head 4 and the servo valve 5 form a hydraulic speaker body 6. [0024] A displacement detector 7 is provided at one end (left end in the drawing) of the piston rod 2 and detects an axial displacement x of the diaphragm 1. Electrical position feedback is applied by the displacement detector 7 so that the diaphragm 1 returns to the home position (center position of vibration). 05-05-2019 5 An acceleration detector 8 is provided at the other end (right end) of the piston rod 2 and detects an output frequency as an acceleration x ′ ′ in the axial direction of the diaphragm 1. [0025] A compensation controller 9 is designed based on the H∞ theory, and performs frequency shaping of a closed loop composed of the hydraulic speaker body 6 and the compensation controller 9. That is, after the intrinsic frequency characteristics generated by the diaphragm 1 driven through the piston rod 2, the hydraulic cylinder 3 and the servo valve 5 are measured by the acceleration detector 8 with respect to the voice input signal r which is known in advance, Compensation controller for compensating the specific frequency characteristic is derived based on H 理論 theory in order to flatten the output frequency characteristic in the target frequency range, and the voice input signal r actually input is derived. The control input signal y is generated based on the compensation controller and the acceleration x ′ ′ to drive the servo valve 5. [0026] The design method of the compensation controller will be described below. The block diagram shown in FIG. 1 (b) represents the hydraulic drive speaker shown in FIG. 1 (a). K (s) represents a compensation controller, and G (s) represents a hydraulic speaker body. FIG. 2 shows the frequency characteristic of the liquid pressure speaker main body, the horizontal axis shows the frequency and the vertical axis shows the gain. [0027] In FIG. 2, the curve L1 (solid line) is the frequency characteristic obtained for the voice input signal that is known in advance, and the curve L2 (dashed dotted line) is the frequency 05-05-2019 6 characteristic of the model based on the physical law of the hydraulic speaker body 6 ( It is given by equation 5). [0028] As represented by the curve L2 shown in the figure, this model expresses the average characteristic of the actually measured frequency characteristic, but the resonance characteristic, particularly the model error in the low region, is large. Therefore, modeling based on measured data is performed. [0030] It can be seen that the frequency characteristic of the measured hydraulic pressure speaker main body 6 is similar to the characteristic inherent to a distributed constant system or a vibration system represented by, for example, a flexible arm or the like. Taking the flexible arm as an example, the transfer function is expressed by Equation 6, which means "rigid body mode + vibration mode". [0032] Therefore, in order to actively introduce the vibration mode, modeling was performed from the second term of the right side of the equation (6). The procedure is shown below. [0033] (1) The resonance frequency is read from the actual measurement data represented by the curve L1 in FIG. 2 to determine the vibration mode of the fluid pressure speaker main body 6 to be controlled. 05-05-2019 7 [0034] (2) Match the average gain over the entire frequency range. [0035] (3) Determine the damping coefficient and gain of each mode according to the vibration mode. [0036] (4) Add dynamics. [0037] (5) Execute (3) again. [0038] (6) Adds no time [0039] As a result of performing the operations shown in the above (1) to (6), a model of Formula 7 is obtained. The frequency characteristic is represented by a curve L3 (broken line) in FIG. [0041] It can be seen from the curve L3 shown in FIG. 2 that the frequency characteristics are reproduced almost completely. Taking this into consideration, the transfer characteristics of this control system can be sufficiently expressed by equation (7). 05-05-2019 8 [0042] Here, H ∞ control is to control the control target so that the value of equation 8 becomes minimum, where V is the transfer function of the closed loop between the signals of interest and WV is its frequency weighting function. is there. [0043] The following control specifications are given to the feedback system shown in FIG. 1 (b). [0044] (Control specification 1) Minimize sensitivity. [0045] By setting the transfer function V to Eq. 9 and minimizing the absolute value (Eq. 10) of the product with the frequency weighting function Ws of Eq. 9, Eq. 9 can be suppressed to a low frequency range. [0046] S (s) = (1 + GK) -1 [0047] (Control specification 2) Robust stability (being robust even with a modeling error) is ensured. [0048] The transfer function V is expressed by Equation 11, and the frequency characteristic is made flat by minimizing the absolute value (Expression 12) of the product with the frequency weighting function WT of Equation 11. [0049] T (s) = GK (1 + GK) -1 [0050] 05-05-2019 9 By the way, finding a compensation controller that satisfies the control specifications 1 and 2 results in the mixed sensitivity problem of H 制 御 control theory with the equation 3 as an evaluation function. be able to. That is, frequency shaping is performed by minimizing the equation (3). [0051] The generalized plant P for this problem is given by Eq. [0053] The problem here is that the controlled object has a zero at the origin (a function representing the frequency characteristic diverges to −∞ at the origin). In that case, the solvable control of the H∞ problem is not satisfied. Therefore, the present inventors have taken the following two measures. [0054] (Countermeasure 1) Eliminate the zero point of the origin by appropriate selection of the weighting function. [0055] (Solution 2) The zero point of the transfer function G (s) is shifted within the range in which the frequency characteristic of the open loop (without feedback) consisting of the compensation controller 9 and the fluid pressure speaker main body 6 does not change. [0056] 05-05-2019 10 As a nominal model obtained by improving the model, a model of equation (14) is used, which is reduced in dimension by removing the fourth-order vibration mode by equation (7) in consideration of the reduction of the order of the compensation controller. Also, note that this nominal model has a zero at the origin, and consider the nominal model of Eq. 15 in which the zero is moved slightly to the right. [0059] When the uncertainty Δ (s) exists in the nominal model, the weighting function WT is selected so as to satisfy the equation (16). [0061] The two weighting functions WT1 (s) and WT2 (s) represented by Eq. 17 and Eq. 18 are selected. [0064] As the weighting function Ws, one is used which suppresses the sensitivity function S (s) uniformly in the frequency range of 10 Hz to 200 Hz. For measure 2, the weighting function is given by Eq. [0066] In the case of measure 1, although the zero point of origin is canceled by the weighting function WT1, it satisfies the solvability with respect to the generalized plant, and the zero can not be canceled essentially. Therefore, it does not make sense with a weight function like Eq. 19 (because S (0) = 0 dB). For measure 1, select Eq. 20 with the weight to the sensitivity function S at the origin kept low. 05-05-2019 11 [0068] From the above, a generalized plant of the equation is obtained and solved by the method of Glover and Doyle. As a result, a compensation controller represented by the following function (Equation 21 to 26) is obtained. [0069] (Measure 1) [0071] (However, Kn1 (s) in the equation 21 satisfies the equation 22 and Kd1 (s) satisfies the equation 23. ） [0074] (Countermeasure 2) [0076] (However, Kn2 (s) in the equation 24 satisfies the equation 25 and Kd2 (s) satisfies the equation 26. ） [0079] The equations (21) and (24) are functions representing the compensation controller, respectively. [0080] The functions represented by the equations (24) to (26) are configured as an integrator and an amplifier as shown in FIG. [0081] 05-05-2019 12 In the figure, 10a to 10h are integrators, and 11a to 11r are amplifiers. The respective coefficients in the equation (25) correspond to the respective amplifiers 11h to 11s, and the respective coefficients in the equation (26) correspond to the respective amplifiers 11a to 11g. The values of the amplifiers of the compensation controller are slightly different because fine adjustment was performed. [0082] FIG. 4 shows the frequency characteristics of this compensation controller, in which the horizontal axis represents frequency and the vertical axis represents gain. As shown in the figure, the frequency characteristic curve shown in FIG. 1 has unevenness in the reverse direction. [0083] Next, the operation of the embodiment will be described. [0084] H∞ The diaphragm acceleration signal corresponding to the sound output from the speaker main body 6 including the diaphragm 1, the piston rod 2, the hydraulic cylinder 3 and the servo valve 5 and the compensation controller 9 designed based on the H 理論 theory, The deviation from the voice input r input to the compensation controller 9 is detected, and the control input signal y is generated based on this deviation and the derived compensation frequency characteristic to drive the servo valve 5, so the hydraulic pressure speaker known in advance After being corrected by the compensation controller 9 having a frequency characteristic opposite to the frequency characteristic of the main body 6, the signal is inputted to the hydraulic speaker main body 6 and the frequency characteristic is shaped into a flat. The diaphragm acceleration corresponding to the sound output generated from the diaphragm 1 based on the shaped frequency characteristic is detected by the acceleration detector 8 and input to the compensation controller 9, and the hydraulic pressure speaker main body 6 based on the deviation thereof. The transfer function is controlled so that the frequency characteristic of the closed loop formed by the compensation controller 9 and the controller 9 becomes flat. 05-05-2019 13 [0085] FIG. 5 shows the frequency characteristic of a closed loop consisting of a compensation controller and a hydraulic pressure speaker, the horizontal axis showing frequency and the vertical axis showing gain. [0086] As shown in the figure, it can be seen that the frequency characteristic of the closed loop formed by the hydraulic speaker body 6 and the compensation controller 9 is flat from 10 Hz to about 300 Hz (the frequency characteristic of the conventional hydraulic speaker body is 10 to about 100 Hz). Further, since position feedback is applied to the diaphragm 1, it is possible to prevent the position of the diaphragm 1 from being biased to one side. [0087] In the above, according to the present embodiment, the compensation controller measures the inherent frequency characteristics generated by the diaphragm with respect to the voice input signal that is known in advance by the acceleration detector, and then the output frequency characteristics in the target frequency range In order to flatten the H 補償 theory, a compensation controller is derived for compensating the specific frequency characteristic, and the derived compensation controller is a control input based on the voice input signal and the acceleration signal that are actually input. Since the signal is generated to drive the servo valve, it is possible to realize a fluid pressure driven speaker capable of producing a sound of a frequency from extremely low frequency to several hundreds Hz. Further, the compensation controller used in the present embodiment is configured of an integrator, an amplifier, and (hardware), but is not limited to this, and may be configured of software. [0088] In the present embodiment, although sensitivity minimization and robust stability are simultaneously performed on the compensation controller, the present invention is not limited to this, and either sensitivity minimization or robust stability may be performed. [0089] As described above, according to the present invention, the following excellent effects are exhibited. [0090] The compensation controller measures the inherent frequency characteristic occurring in the diaphragm with respect to the voice input signal which is known in advance by the acceleration detector, and then the compensation controller for compensating the inherent frequency characteristic based on the H∞ theory. The derived and derived compensation controller generates the control input signal based on the voice input signal and the acceleration 05-05-2019 14 signal that are actually input to drive the servo valve, so that the frequency from extremely low frequency to several hundred Hz is It is possible to realize a hydraulic drive type speaker that can make a sound. 05-05-2019 15

1/--страниц