A method and device for identifying parameters of a linear resonant motor

Abstract

The embodiment of the present invention provides a method and device for linear resonant motor parameter identification. The LRA parameter identification uses the zero-input response discrete-time system model of the second-order linear resonant system to construct a second-order forward prediction error filter, and uses the augmented Wienerhof equation to solve the coefficients of the discrete-time system model to obtain the minimum mean square error estimate, which has the advantages of strong anti-noise ability and good robustness. Through the estimated discrete-time system model, the impulse response invariant method can directly calculate the poles, free oscillation frequency and damping coefficient of the continuous-time system without iterative calculations.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of drive circuit technology, and specifically to a method and device for identifying parameters of a linear resonant motor. Background Art

[0002] Linear resonant actuator (LRA) is usually used to provide tactile feedback effects on portable terminals. LRA includes components such as springs, coils and vibrators. It is driven by the LRA driver chip. The driver chip applies an excitation current to the coil to generate a magnetic field, which pushes the magnetic vibrator to move in a certain direction. When the direction of the excitation current changes, the magnetic field and the driving force also change. Therefore, if the driver chip applies a periodic voltage signal to the coil, the periodic excitation current generated by it will push the vibrator to vibrate back and forth, achieving the effect of tactile feedback. Due to the resonant characteristics of LRA, the amplitude of the vibrator vibration shows a bandpass characteristic with the driving signal frequency. When the driving signal frequency is at the natural frequency of the vibrator (F0), the amplitude of the vibrator vibration reaches the highest and the vibration efficiency is the best.

[0003] The prior art detects the zero-crossing point of the back electromagnetic flux (BEMF) waveform when the driving voltage waveform is turned off, and calculates the damped oscillation frequency of the LRA by averaging the zero-crossing interval. This solution has the following disadvantages:

[0004] (1) There are high requirements for the accuracy of the BEMF detection circuit. Since the BEMF waveform is a damped oscillation waveform, as the number of cycles increases, the BEMF amplitude will gradually decay to below the noise threshold. Under low signal-to-noise ratio, the zero-crossing detection will be affected by noise interference and produce jitter. Even if it is averaged several times, it is still difficult to ensure accuracy.

[0005] (2) The existing technology only samples data at the BEMF zero crossing point or peak point, and loses the sample points at other positions of the time domain waveform, resulting in information loss, and thus poor noise resistance. During the BEMF detection process, if the portable device vibrates due to external force and generates noise, the identified damped oscillation frequency will have a large error.

[0006] (3) There is a relationship between the damped oscillation frequency and the free oscillation frequency of the second-order linear resonant system. The error is ξ, which is the motor damping coefficient. The damping coefficient needs to be measured and calibrated to get the free oscillation frequency. The damping coefficient needs to be calculated by the attenuation coefficient of the envelope of the BEMF during the damped oscillation process and the free oscillation frequency. Therefore, under the condition that only the damped oscillation frequency and the envelope attenuation coefficient are obtained by measurement, the damping coefficient ξ and the free oscillation frequency need to be calculated by iteration. Summary of the invention

[0007] To this end, an embodiment of the present invention provides a linear resonant motor parameter identification method and device to solve the technical problems in the prior art.

[0008] In order to achieve the above purpose, the embodiment of the present invention provides the following technical solutions:

[0009] According to a first aspect of an embodiment of the present invention, an embodiment of the present application provides a method for identifying parameters of a linear resonant motor, the method comprising:

[0010] According to the amplitude input, a periodic driving waveform signal of a preset frequency is generated for driving the linear resonant motor through a driving circuit;

[0011] When the driving waveform is played for several cycles, the linear resonant motor obtains a non-zero initial displacement and initial speed, and the driving circuit is turned off;

[0012] Collecting a plurality of cycles of voltage waveforms at both ends of the linear resonant motor to obtain a reverse electromotive force collection sequence x[n] of the linear resonant motor at a predetermined number of sampling points, wherein n is greater than or equal to 2;

[0013] The back electromotive force acquisition sequence x[n] is used to perform second-order forward prediction error processing to obtain the minimum mean square error estimation value a of the first tap coefficient and the second tap coefficient. 1 and a 2 ;

[0014] According to the minimum mean square error estimation value a of the first tap coefficient and the second tap coefficient 1 and a 2 Perform parameter identification and obtain the free oscillation frequency f n and the damping coefficient ξ.

[0015] Furthermore, the back electromotive force acquisition sequence x[n] is used to perform second-order forward prediction error processing, including:

[0016] Using the back electromotive force acquisition sequence x[n] to perform a delay process for one sampling period, to obtain a first delayed back electromotive force sequence x[n-1];

[0017] Using the first delayed back electromotive force sequence x[n-1] to perform a delay process for one sampling period, a second delayed back electromotive force sequence x[n-2] is obtained;

[0018] The back electromotive force acquisition sequence x[n], the first delayed back electromotive force sequence x[n-1], and the second delayed back electromotive force sequence x[n-2] are used to calculate and obtain the back electromotive force output sequence f 2[n], the reverse electromotive force output sequence f 2 The calculation formula for [n] is as follows:

[0019] f 2 [n]=x[n]+a 1 ·x[n-1]+a 2 x[n-2];

[0020] The reverse electromotive force output sequence f 2 [n] Calculate the biased estimate of the autocorrelation function r(k) delayed by k samples Biased Estimation The calculation formula is as follows:

[0021]

[0022] Where N is the reverse electromotive force output sequence f 2 The length of [n], k = 0, 1, 2;

[0023] Using biased estimates The calculation results construct the following augmented Wienerhof equation:

[0024]

[0025] Among them, P 2 is the reverse electromotive force output sequence f 2 The mean square value of [n];

[0026] Solve the augmented Wienerhof equation to obtain the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient. 1 and a 2 .

[0027] Further, according to the minimum mean square error estimation value a of the first tap coefficient and the second tap coefficient 1 and a 2 Perform parameter identification processing, including:

[0028] The minimum mean square error estimate a of the first tap coefficient and the second tap coefficient is used. 1 and a 2 , solve the characteristic equation of the following second-order discrete time system and obtain the two roots p of the characteristic equation d1 、p d2 :

[0029] 1+a 1 z -1 +a 2 z -2 =0

[0030] Where z is the complex variable of the characteristic equation of the second-order discrete-time system;

[0031] Using the two roots p of the characteristic equation d1 、p d2 , calculate the two pole values ​​p of the second-order continuous-time system 1 、p 2 , two extreme values ​​p 1 、p 2 The calculation formulas are as follows:

[0032] p 1 =lnp d1 / T

[0033] p 2 =lnp d2 / T

[0034] Where, T is the sampling period of the back EMF acquisition sequence;

[0035] Using two extreme values ​​p 1 、p 2 Calculate the free oscillation frequency f n and damping coefficient ξ, the free oscillation frequency f n And the calculation formula of the damping coefficient ξ is as follows:

[0036]

[0037] As a preferred embodiment of the present application, the method further includes:

[0038] Determine the free oscillation frequency f n Whether it is between the preset upper and lower thresholds of the free oscillation frequency;

[0039] If the free oscillation frequency f n If the frequency is not between the preset upper and lower thresholds of the free oscillation frequency, the free oscillation frequency f will be reported to the upper computer. n The error signal indicates that the linear resonant motor is defective or damaged.

[0040] As a preferred embodiment of the present application, the method further includes:

[0041] If the free oscillation frequency f n If the free oscillation frequency is between the upper and lower thresholds, it is determined whether the damping coefficient ξ is between the upper and lower thresholds;

[0042] If the damping coefficient ξ is not between the preset upper and lower thresholds of the damping coefficient, a signal indicating that the damping coefficient ξ is wrong is reported to the upper computer, indicating that the linear resonance motor is defective or damaged;

[0043] If the damping coefficient ξ is between the preset upper and lower thresholds of the damping coefficient, a signal indicating that the damping coefficient ξ is correct is reported to the host computer, indicating that the linear resonance motor is good or not damaged.

[0044] As a preferred embodiment of the present application, the method further includes:

[0045] If the linear resonant motor is good or not damaged, the free oscillation frequency f is calculated. n , when the driving circuit drives the linear motor next time, the playback period T of the periodic driving waveform signal is determined drv , the next play cycle T drv The calculation formula is as follows:

[0046] T drv =1 / f n .

[0047] As a preferred embodiment of the present application, the method further includes:

[0048] By changing the sampling rate of the driving voltage waveform, the frequency of the periodic driving voltage signal of the existing linear resonant motor is adjusted to the free oscillation frequency f of the linear resonant motor. n , to maximize the vibration amplitude and efficiency of the linear resonant motor.

[0049] According to a second aspect of an embodiment of the present invention, an embodiment of the present application provides a linear resonant motor parameter identification system, the system comprising:

[0050] A driving waveform generating module, used for generating a periodic driving waveform signal of a preset frequency according to the amplitude input size, and outputting the periodic driving waveform signal to a driving circuit to drive the linear resonant motor;

[0051] The analog-to-digital conversion circuit is used to, when the driving waveform is played for several cycles, the linear resonant motor obtains a non-zero initial displacement and initial speed, and the driving circuit is turned off; the voltage waveforms at both ends of the linear resonant motor for several cycles are collected to obtain a reverse electromotive force collection sequence x[n] of the linear resonant motor at a predetermined number of sampling points, wherein n is greater than or equal to 2;

[0052] A second-order forward prediction error filter is used to perform second-order forward prediction error processing using the back electromotive force acquisition sequence x[n] to obtain a minimum mean square error estimate a of the first tap coefficient and the second tap coefficient. 1 and a 2 ;

[0053] A parameter identification module for determining the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient1 and a 2 Perform parameter identification and obtain the free oscillation frequency f n and the damping coefficient ξ.

[0054] Furthermore, the back electromotive force acquisition sequence x[n] is used to perform second-order forward prediction error processing, including:

[0055] Using the back electromotive force acquisition sequence x[n] to perform a delay process for one sampling period, to obtain a first delayed back electromotive force sequence x[n-1];

[0056] Using the first delayed back electromotive force sequence x[n-1] to perform a delay process for one sampling period, a second delayed back electromotive force sequence x[n-2] is obtained;

[0057] The back electromotive force acquisition sequence x[n], the first delayed back electromotive force sequence x[n-1], and the second delayed back electromotive force sequence x[n-2] are used to calculate and obtain the back electromotive force output sequence f 2 [n], the reverse electromotive force output sequence f 2 The calculation formula for [n] is as follows:

[0058] f 2 [n]=x[n]+a 1 ·x[n-1]+a 2 x[n-2];

[0059] The reverse electromotive force output sequence f 2 [n] Calculate the biased estimate of the autocorrelation function r(k) delayed by k samples Biased Estimation The calculation formula is as follows:

[0060]

[0061] Where N is the reverse electromotive force output sequence f 2 The length of [n], k = 0, 1, 2;

[0062] Using biased estimates The calculation results construct the following augmented Wienerhof equation:

[0063]

[0064] Among them, P 2 is the reverse electromotive force output sequence f 2 The mean square value of [n];

[0065] Solve the augmented Wienerhof equation to obtain the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient.1 and a 2 .

[0066] Further, according to the minimum mean square error estimation value a of the first tap coefficient and the second tap coefficient 1 and a 2 Perform parameter identification processing, including:

[0067] The minimum mean square error estimate a of the first tap coefficient and the second tap coefficient is used. 1 and a 2 , solve the characteristic equation of the following second-order discrete time system and obtain the two roots p of the characteristic equation d1 、pd 2 :

[0068] 1+a 1 z -1 +a 2 z -2 =0

[0069] Where z is the complex variable of the characteristic equation of the second-order discrete-time system;

[0070] Using the two roots p of the characteristic equation d1 、p d2 , calculate the two pole values ​​p of the second-order continuous-time system 1 、p 2 , two extreme values ​​p 1 、p 2 The calculation formulas are as follows:

[0071] p 1 =lnp d1 / T

[0072] p 2 =lnp d2 / T

[0073] Where, T is the sampling period of the back EMF acquisition sequence;

[0074] Using two extreme values ​​p 1 、p 2 Calculate the free oscillation frequency f n and damping coefficient ξ, the free oscillation frequency f n And the calculation formula of the damping coefficient ξ is as follows:

[0075]

[0076] As a preferred embodiment of the present application, the parameter identification module is also used for:

[0077] Determine the free oscillation frequency f nWhether it is between the preset upper and lower thresholds of the free oscillation frequency;

[0078] If the free oscillation frequency f n If the frequency is not between the preset upper and lower thresholds of the free oscillation frequency, the free oscillation frequency f will be reported to the upper computer. n An error signal indicates that the linear resonant motor is defective or damaged;

[0079] If the free oscillation frequency f n If the free oscillation frequency is between the upper and lower thresholds, it is determined whether the damping coefficient ξ is between the upper and lower thresholds;

[0080] If the damping coefficient ξ is not between the preset upper and lower thresholds of the damping coefficient, a signal indicating that the damping coefficient ξ is wrong is reported to the upper computer, indicating that the linear resonance motor is defective or damaged;

[0081] If the damping coefficient ξ is between the preset upper and lower thresholds of the damping coefficient, a signal indicating that the damping coefficient ξ is correct is reported to the host computer, indicating that the linear resonance motor is good or not damaged.

[0082] As a preferred embodiment of the present application, the parameter identification module is also used for:

[0083] If the linear resonant motor is good or not damaged, the free oscillation frequency f n Sent to the driving waveform generation module to use the calculated free oscillation frequency f n , when the driving circuit drives the linear motor next time, the playback period T of the periodic driving waveform signal is determined drv , the next play cycle T drv The calculation formula is as follows:

[0084] T drv =1 / f n ;

[0085] The free oscillation frequency f n The drive waveform generation module is sent to adjust the frequency of the periodic drive voltage signal of the existing linear resonant motor to the free oscillation frequency f of the linear resonant motor by changing the sampling rate of the drive voltage waveform. n , to maximize the vibration amplitude and efficiency of the linear resonant motor.

[0086] Compared with the prior art, the linear resonant motor parameter identification method and system provided by the embodiment of the present application uses the zero-input response discrete-time system model of the second-order linear resonant system to construct a second-order forward prediction error filter, and uses the augmented Wienerhof equation to solve the coefficients of the discrete-time system model. The minimum mean square error estimation has the advantages of strong anti-noise ability and good robustness. Through the estimated discrete-time system model, the poles, free oscillation frequency and damping coefficient of the continuous-time system can be directly calculated by the impulse response invariance method without iterative calculations. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0088] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0089] Figure 1 A schematic diagram of the structure of a linear resonant motor parameter identification system provided by an embodiment of the present invention;

[0090] Figure 2 A schematic flow chart of a linear resonant motor parameter identification method provided by an embodiment of the present invention;

[0091] Figure 3 A schematic diagram of the voltage waveform at both ends of the LRA and the time position of the collected data in a linear resonance motor parameter identification method provided by an embodiment of the present invention;

[0092] Figure 4 The back electromotive force output sequence f is obtained in the second-order forward prediction error processing in a linear resonant motor parameter identification method provided by an embodiment of the present invention. 2 Schematic diagram of the principle of [n]. DETAILED DESCRIPTION

[0093] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0094] The linear resonant motor parameters that need to be identified in the embodiment of the present invention include: the free oscillation angular frequency and the damping coefficient in the transfer function from the driving voltage to the back electromagnetic flux (BEMF) of the linear resonant motor.

[0095] In the existing detection method, the driver chip turns off the driving signal in the middle of providing the driving voltage waveform, allowing the LRA oscillator to oscillate freely for several cycles, and uses the acquisition circuit to collect the BEMF generated when the oscillator oscillates by itself. BEMF is a damped oscillation waveform. The damped oscillation frequency of the LRA is calculated by averaging the zero-crossing intervals of the induced electromotive force waveform and taking the inverse. Since the BEMF waveform is a damped oscillation waveform, as the number of cycles increases, the BEMF amplitude will gradually decay below the noise threshold. Zero-crossing detection under low signal-to-noise ratio will be disturbed by noise and produce jitter. Even if it is averaged several times, it is still difficult to ensure accuracy. The existing technology only samples data at the BEMF zero-crossing point or peak point, and loses the sample points at other positions of the time domain waveform, resulting in information loss, so the anti-noise ability is poor. In the process of detecting BEMF, if the portable device vibrates due to external force and generates noise, the damped oscillation frequency identified at this time will have a large error. The measurement of the damping coefficient is obtained by iterative calculation by detecting the attenuation coefficient of the envelope of the BEMF during the damped oscillation process and the damped oscillation frequency.

[0096] In order to solve the above technical problems, Figure 1 As shown, an embodiment of the present application provides a linear resonant motor parameter identification system, the system comprising: a drive waveform generation module 01, an analog-to-digital conversion circuit 02, a second-order forward prediction error filter 03, and a parameter identification module 04.

[0097] Specifically, the driving waveform generation module 01 is used to generate a periodic driving waveform signal of a preset frequency according to the amplitude input size, and output the periodic driving waveform signal to the driving circuit 05 to drive the linear resonant motor 06; in the embodiment of the present invention, two driving circuits 05 drive a linear resonant motor 06 together; the analog-to-digital conversion circuit 02 is used to drive the linear resonant motor 06 to obtain a non-zero initial displacement and initial speed when the driving waveform is played for several cycles, and turn off the driving circuit 05; collect the voltage waveforms at both ends of the linear resonant motor 06 for several cycles to obtain the reverse electromotive force acquisition sequence x[n] of the linear resonant motor 06 at a predetermined number of sampling points, where n is greater than or equal to 2; the second-order forward prediction error filter 03 is used to use the reverse electromotive force acquisition sequence x[n] to perform second-order forward prediction error processing to obtain the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient. 1 and a 2 ; Parameter identification module 04 is used to estimate the minimum mean square error of the first tap coefficient and the second tap coefficient a 1 and a 2 Perform parameter identification and obtain the free oscillation frequency f n and the damping coefficient ξ.

[0098] Compared with the prior art, the linear resonant motor parameter identification method and system provided by the embodiment of the present application uses the zero-input response discrete-time system model of the second-order linear resonant system to construct a second-order forward prediction error filter, and uses the augmented Wienerhof equation to solve the coefficients of the discrete-time system model. The minimum mean square error estimation has the advantages of strong anti-noise ability and good robustness. Through the estimated discrete-time system model, the poles, free oscillation frequency and damping coefficient of the continuous-time system can be directly calculated by the impulse response invariance method without iterative calculations.

[0099] Corresponding to the above-disclosed linear resonant motor parameter identification system, the embodiment of the present invention further discloses a linear resonant motor parameter identification method. The following describes in detail a linear resonant motor parameter identification method disclosed in the embodiment of the present invention in combination with the above-described linear resonant motor parameter identification system.

[0100] like Figure 2As shown, an embodiment of the present application provides a method for identifying parameters of a linear resonant motor, the method comprising: a driving waveform generating module 01 generates a periodic driving waveform signal of a preset frequency according to the amplitude input size, and outputs the periodic driving waveform signal to a driving circuit 05 to drive a linear resonant motor 06; in an embodiment of the present invention, two driving circuits 05 drive a linear resonant motor 06 together; when the driving waveform plays a number of cycles, the linear resonant motor 06 obtains a non-zero initial displacement and initial speed, and the driving circuit 05 is turned off; an analog-to-digital converter circuit 02 collects a number of cycles of voltage waveforms at both ends of the linear resonant motor 06 to obtain a reverse electromotive force collection sequence x[n] of the linear resonant motor 06 at a predetermined number of sampling points, wherein n is greater than or equal to 2, and the reverse electromotive force collection sequence x[n] is output to a second-order forward prediction error filter 03, as shown in FIG. Figure 3 As shown; the second-order forward prediction error filter 03 uses the back electromotive force acquisition sequence x[n] to perform second-order forward prediction error processing to obtain the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient 1 and a 2 ; Parameter identification module 04 is based on the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient 1 and a 2 Perform parameter identification and obtain the free oscillation frequency f n and the damping coefficient ξ.

[0101] Specifically, the second-order forward prediction error filter 03 uses the back electromotive force acquisition sequence x[n] to perform second-order forward prediction error processing, including: using the back electromotive force acquisition sequence x[n] to perform a delay process for one sampling cycle to obtain a first delayed back electromotive force sequence x[n-1]; using the first delayed back electromotive force sequence x[n-1] to perform a delay process for one sampling cycle to obtain a second delayed back electromotive force sequence x[n-2]; using the back electromotive force acquisition sequence x[n], the first delayed back electromotive force sequence x[n-1], and the second delayed back electromotive force sequence x[n-2] to calculate and obtain a back electromotive force output sequence f 2 [n], the reverse electromotive force output sequence f 2 The calculation formula for [n] is as follows:

[0102] f 2 [n]=x[n]+a 1 ·x[n-1]+a 2 x[n-2];

[0103] The above content is specifically referenced Figure 4, and then the reverse electromotive force output sequence f 2 [n] Calculate the biased estimate of the autocorrelation function r(k) delayed by k samples Biased Estimation The calculation formula is as follows:

[0104]

[0105] Where N is the reverse electromotive force output sequence f 2 The length of [n], k = 0, 1, 2;

[0106] Using biased estimates The calculation results construct the following augmented Wienerhof equation:

[0107]

[0108] Among them, P 2 is the reverse electromotive force output sequence f 2 The mean square value of [n];

[0109] Solve the augmented Wienerhof equation to obtain the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient. 1 and a 2 .

[0110] Further, the parameter identification module 04 estimates the minimum mean square error of the first tap coefficient and the second tap coefficient according to a 1 and a 2 Performing parameter identification processing, specifically comprising: using the minimum mean square error estimation value a of the first tap coefficient and the second tap coefficient 1 and a 2 , solve the characteristic equation of the following second-order discrete time system and obtain the two roots p of the characteristic equation d1 、p d2 :

[0111] 1+a 1 z -1 +a 2 z -2 =0

[0112] Among them, z -1 It means that the back electromotive force acquisition sequence x[n] is delayed by one sampling cycle, that is, z is the complex variable of the characteristic equation of the second-order discrete time system;

[0113] Using the two roots p of the characteristic equation d1 、p d2 , calculate the two pole values ​​p of the second-order continuous-time system 1 、p 2, two extreme values ​​p 1 、p 2 The calculation formulas are as follows:

[0114] p 1 =lnp d1 / T

[0115] p 2 =lnp d2 / T

[0116] Where, T is the sampling period of the back EMF acquisition sequence;

[0117] Using two extreme values ​​p 1 、p 2 Calculate the free oscillation frequency f n and damping coefficient ξ, the free oscillation frequency f n And the calculation formula of the damping coefficient ξ is as follows:

[0118]

[0119] It should be noted that the driving circuit 01 drives and adjusts the vibration speed and amplitude of the motor 05 according to the adjusted driving waveform amplitude and period, so that the linear resonance motor 05 can be controlled with higher precision and speed.

[0120] refer to Figure 2 Preferably, a linear resonant motor parameter identification method disclosed in an embodiment of the present invention further includes: the parameter identification module 04 is also used to determine the free oscillation frequency f n Is it between the preset upper and lower thresholds of the free oscillation frequency? If the free oscillation frequency f n If the frequency is not between the preset upper and lower limits of the free oscillation frequency, a signal indicating that the linear resonant motor parameters are incorrect will be reported to the upper computer, that is, the free oscillation frequency f will be reported to the upper computer. n The error signal indicates that the linear resonant motor is defective or damaged; if the free oscillation frequency f n The preset free oscillation frequency upper and lower thresholds (f n，lower_limit , f n，upper_limit ), then determine whether the damping coefficient ξ is within the preset damping coefficient upper and lower thresholds (ξ lower_limit ,ξ upper_limit ); if the damping coefficient ξ is not between the preset upper and lower thresholds of the damping coefficient, a signal that the linear resonance motor parameters are wrong is reported to the host computer, that is, a signal that the damping coefficient ξ is wrong is reported to the host computer, indicating that the linear resonance motor is defective or damaged; if the damping coefficient ξ is between the preset upper and lower thresholds of the damping coefficient, a signal that the damping coefficient ξ is correct is reported to the host computer, indicating that the linear resonance motor is good or not damaged.

[0121] refer to Figure 2 Preferably, a linear resonant motor parameter identification method disclosed in an embodiment of the present invention further includes: if the linear resonant motor is good or not damaged, the parameter identification module 04 is also used to convert the free oscillation frequency f n The free oscillation frequency f is calculated by the driving waveform generating module 01. n , when the driving circuit drives the linear motor next time, the playback period T of the periodic driving waveform signal is determined drv , the next play cycle T drv The calculation formula is as follows:

[0122] T drv =1 / f n ;

[0123] The parameter identification module 04 is also used to convert the free oscillation frequency f n The driving waveform generating module 01 adjusts the periodic driving voltage signal frequency of the existing linear resonant motor to the free oscillation frequency f of the linear resonant motor by changing the sampling rate of the driving voltage waveform. n , to maximize the vibration amplitude and efficiency of the linear resonant motor.

[0124] Compared with the prior art, the linear resonant motor parameter identification method and system provided by the embodiment of the present application uses the zero-input response discrete-time system model of the second-order linear resonant system to construct a second-order forward prediction error filter, and uses the augmented Wienerhof equation to solve the coefficients of the discrete-time system model. The minimum mean square error estimation has the advantages of strong anti-noise ability and good robustness. Through the estimated discrete-time system model, the poles, free oscillation frequency and damping coefficient of the continuous-time system can be directly calculated by the impulse response invariance method without iterative calculations.

[0125] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A method for identifying parameters of a linear resonant motor, It is characterized in that The method comprises: According to the amplitude input, a periodic driving waveform signal of a preset frequency is generated for driving the linear resonant motor through a driving circuit; When the driving waveform is played for several cycles, the linear resonant motor obtains a non-zero initial displacement and initial speed, and the driving circuit is turned off; Collecting a plurality of cycles of voltage waveforms at both ends of the linear resonant motor to obtain a reverse electromotive force collection sequence x[n] of the linear resonant motor at a predetermined number of sampling points, wherein n is greater than or equal to 2; The back electromotive force acquisition sequence x[n] is used to perform second-order forward prediction error processing to obtain the minimum mean square error estimation value a of the first tap coefficient and the second tap coefficient. 1 and a 2 ; According to the minimum mean square error estimation value a of the first tap coefficient and the second tap coefficient 1 and a 2 Perform parameter identification and obtain the free oscillation frequency f n and damping coefficient ξ; The back electromotive force acquisition sequence x[n] is used to perform second-order forward prediction error processing, including: Using the back electromotive force acquisition sequence x[n] to perform a delay process for one sampling period, to obtain a first delayed back electromotive force sequence x[n-1]; Using the first delayed back electromotive force sequence x[n-1] to perform a delay process for one sampling period, a second delayed back electromotive force sequence x[n-2] is obtained; The back electromotive force acquisition sequence x[n], the first delayed back electromotive force sequence x[n-1], and the second delayed back electromotive force sequence x[n-2] are used to calculate and obtain the back electromotive force output sequence f 2 [n], the reverse electromotive force output sequence f 2 The calculation formula for [n] is as follows: f 2 [n]=x[n]+a 1 ·x[n-1]+a 2 ·x[n-2]; The reverse electromotive force output sequence f 2 [n] Calculate the biased estimate of the autocorrelation function r(k) delayed by k samples Biased Estimation The calculation formula is as follows: Where N is the reverse electromotive force output sequence f 2 The length of [n], k = 0, 1, 2; Using biased estimates The calculation results construct the following augmented Wienerhof equation: Among them, P 2 is the reverse electromotive force output sequence f 2 The mean square value of [n]; Solve the augmented Wienerhof equation to obtain the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient. 1 and a 2 .

2. A linear resonant motor parameter identification method as claimed in claim 1, It is characterized in that According to the minimum mean square error estimation value a of the first tap coefficient and the second tap coefficient 1 and a 2 Perform parameter identification processing, including: The minimum mean square error estimate a of the first tap coefficient and the second tap coefficient is used. 1 and a 2 , solve the characteristic equation of the following second-order discrete-time system and obtain the two roots pd of the characteristic equation 1 、pd2: 1+a 1 from -1 +a 2 from -2 =0 Where z is the complex variable of the characteristic equation of the second-order discrete-time system; Using the two roots p of the characteristic equation d1 、p d2 , calculate the two pole values ​​p of the second-order continuous-time system 1 、p 2 , two extreme values ​​p 1 、p 2 The calculation formulas are as follows: p 1 =lnp d1 / T p 2 =lnp d2 / T Where, T is the sampling period of the back EMF acquisition sequence; Using two extreme values ​​p 1 、p 2 Calculate the free oscillation frequency f n and damping coefficient ξ, the free oscillation frequency f n And the calculation formula of the damping coefficient ξ is as follows:

3. A linear resonant motor parameter identification method according to any one of claims 1 to 2, It is characterized in that The method further comprises: Determine the free oscillation frequency f n Whether it is between the preset upper and lower thresholds of the free oscillation frequency; If the free oscillation frequency f n If the frequency is not between the preset upper and lower limits of the free oscillation frequency, the free oscillation frequency f will be reported to the upper computer. n The error signal indicates that the linear resonant motor is defective or damaged.

4. A linear resonant motor parameter identification method as claimed in claim 3, It is characterized in that The method further comprises: If the free oscillation frequency f n If the free oscillation frequency is between the preset upper and lower thresholds, it is determined whether the damping coefficient ξ is between the preset upper and lower thresholds of the damping coefficient; If the damping coefficient ξ is not between the preset upper and lower thresholds of the damping coefficient, a signal indicating that the damping coefficient ξ is wrong is reported to the upper computer, indicating that the linear resonance motor is defective or damaged; If the damping coefficient ξ is between the preset upper and lower thresholds of the damping coefficient, a signal indicating that the damping coefficient ξ is correct is reported to the host computer, indicating that the linear resonance motor is good or not damaged.

5. A linear resonant motor parameter identification method as claimed in claim 4, It is characterized in that The method further comprises: If the linear resonant motor is good or not damaged, the free oscillation frequency f is calculated. n , when the driving circuit drives the linear motor next time, the playback period T of the periodic driving waveform signal is determined drv , the next play cycle T drv The calculation formula is as follows: T drv =1 / f n 。 6. A linear resonant motor parameter identification method as claimed in claim 5, It is characterized in that The method further comprises: By changing the sampling rate of the driving voltage waveform, the frequency of the periodic driving voltage signal of the existing linear resonant motor is adjusted to the free oscillation frequency f of the linear resonant motor. n , to maximize the vibration amplitude and efficiency of the linear resonant motor.

7. A linear resonant motor parameter identification system, It is characterized in that The system comprises: A driving waveform generating module, used for generating a periodic driving waveform signal of a preset frequency according to the amplitude input size, and outputting the periodic driving waveform signal to a driving circuit to drive the linear resonant motor; The analog-to-digital conversion circuit is used to, when the driving waveform is played for several cycles, the linear resonant motor obtains a non-zero initial displacement and initial speed, and the driving circuit is turned off; the voltage waveforms at both ends of the linear resonant motor for several cycles are collected to obtain a reverse electromotive force collection sequence x[n] of the linear resonant motor at a predetermined number of sampling points, wherein n is greater than or equal to 2; A second-order forward prediction error filter is used to perform second-order forward prediction error processing using the back electromotive force acquisition sequence x[n] to obtain a minimum mean square error estimate a of the first tap coefficient and the second tap coefficient. 1 and a 2 ; A parameter identification module for determining the minimum mean square error estimate a of the first tap coefficient and the second tap coefficient 1 and a 2 Perform parameter identification and obtain the free oscillation frequency f n and the damping coefficient ξ.

8. A linear resonant motor parameter identification system as claimed in claim 7, It is characterized in that The parameter identification module is also used for: Determine the free oscillation frequency f n Whether it is between the preset upper and lower thresholds of the free oscillation frequency; If the free oscillation frequency f n If the frequency is not between the preset upper and lower limits of the free oscillation frequency, the free oscillation frequency f will be reported to the upper computer. n An error signal indicates that the linear resonant motor is defective or damaged; If the free oscillation frequency f n If the free oscillation frequency is between the preset upper and lower thresholds, it is determined whether the damping coefficient ξ is between the preset upper and lower thresholds of the damping coefficient; If the damping coefficient ξ is not between the preset upper and lower thresholds of the damping coefficient, a signal indicating that the damping coefficient ξ is wrong is reported to the upper computer, indicating that the linear resonance motor is defective or damaged; If the damping coefficient ξ is between the preset upper and lower thresholds of the damping coefficient, a signal indicating that the damping coefficient ξ is correct is reported to the host computer, indicating that the linear resonance motor is good or not damaged.

9. A linear resonant motor parameter identification system as claimed in claim 7, It is characterized in that The parameter identification module is also used for: If the linear resonant motor is good or not damaged, the free oscillation frequency f n Sent to the driving waveform generation module to use the calculated free oscillation frequency f n , when the driving circuit drives the linear motor next time, the playback period T of the periodic driving waveform signal is determined drv , the next play cycle T drv The calculation formula is as follows: T drv =1 / f n ; The free oscillation frequency f n The drive waveform generation module is sent to adjust the frequency of the periodic drive voltage signal of the existing linear resonant motor to the free oscillation frequency f of the linear resonant motor by changing the sampling rate of the drive voltage waveform. n , to maximize the vibration amplitude and efficiency of the linear resonant motor.

Images