Segmental characterization of electromechanical actuators

Abstract

The method includes applying a high frequency signal to an electromechanical actuator and measuring a first response of the electromechanical actuator to the high frequency signal, and estimating electrical parameters of the electromechanical actuator based on the first response; applying a low frequency wide band signal to the electromechanical actuator and measuring a second response of the electromechanical actuator to the low frequency wide band signal, and estimating mechanical parameters of the electromechanical actuator based on the second response and the estimated electrical parameters.

Description

[Background technology]

[0001] (background) Electromechanical actuators typically require calibration, which involves characterizing the actuator by stimulating the actuator. Characterizing the actuator involves determining the actuator's parameters. One known method of stimulating an electromechanical actuator to characterize it is to use a chirp signal. A chirp signal is a sinusoidal signal whose frequency increases or decreases over time to sweep through many different frequencies. However, a drawback of using a chirp signal to characterize an electromechanical actuator is the relatively long amount of time required to obtain accurate results, which may be undesirable in some situations, such as during the manufacture of a large number of devices that include electromechanical actuators that need to be calibrated. Another drawback of using a chirp signal to characterize an electromechanical actuator is that the chirp signal is perceptible. Yet another drawback of using a chirp signal to characterize an electromechanical actuator is that the chirp signal may induce thermal effects, such as heating, if applied for a long enough period of time.

[0002] Another known method of stimulating an electromechanical actuator to characterize it is to use a short-duration, high-amplitude impulse and measure the impulse response. However, a drawback of using an impulse to characterize an electromechanical actuator is that the impulse can be destructive to the electromechanical actuator. Furthermore, there are situations in which using an impulse to characterize an electromechanical actuator is not practically feasible.

[0003] Generally speaking, it would be desirable to have a method for characterizing electromechanical actuators that is fast, accurate, reproducible, and imperceptible. Summary of the Invention [Means for solving the problem]

[0004] (summary) In one embodiment, the present disclosure presents a method, the method including applying a high-frequency signal to an electromechanical actuator, measuring a first response of the electromechanical actuator to the high-frequency signal, and estimating electrical parameters of the electromechanical actuator based on the first response; applying a low-frequency wideband signal to the electromechanical actuator, measuring a second response of the electromechanical actuator to the low-frequency wideband signal, and estimating mechanical parameters of the electromechanical actuator based on the second response and the estimated electrical parameters.

[0005] In another embodiment, the present disclosure provides a non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing a method, the method including applying a high-frequency signal to an electromechanical actuator and measuring a first response of the electromechanical actuator to the high-frequency signal and estimating an electrical parameter of the electromechanical actuator based on the first response; applying a low-frequency wideband signal to the electromechanical actuator and measuring a second response of the electromechanical actuator to the low-frequency wideband signal and estimating a mechanical parameter of the electromechanical actuator based on the second response and the estimated electrical parameter. [Brief explanation of the drawings]

[0006] [Figure 1] FIG. 1 is an exemplary circuit diagram characterization of an electromechanical actuator electrical impedance whose parameters can be estimated in a piecewise manner according to an embodiment of the present disclosure.

[0007] [Figure 2] FIG. 2 is an exemplary diagram illustrating a process for performing piecewise estimation of parameters of an electromechanical actuator, such as the electromechanical actuator of FIG. 1, according to an embodiment of the present disclosure.

[0008] [Figure 3] FIG. 3 is an exemplary diagram illustrating a first step of a piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure.

[0009] [Figure 4] FIG. 4 is an exemplary diagram illustrating high-frequency compensation of low-frequency estimates employed in the first step of the piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure.

[0010] [Figure 5] FIG. 5 is an example graph of the magnitude of the impedance of the tank of an electromechanical actuator as a function of frequency, according to an embodiment of the present disclosure.

[0011] [Figure 6] FIG. 6 is an exemplary individual time and frequency domain graph of an exemplary LF broadband excitation signal employed in the second step of the piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure.

[0012] [Figure 7] FIG. 7 is an exemplary diagram illustrating a second step of a piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure.

[0013] [Figure 8] FIG. 8 is an exemplary graph of HF tone and LF broadband excitation signals employed in a piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0014] (Detailed explanation) FIG. 1 is an example circuit diagram characterization of the electrical impedance of an electromechanical actuator 100, whose parameters can be estimated in a piecewise manner, according to an embodiment of the present disclosure. The electromechanical actuator 100 may be a voice coil motor, a linear resonant actuator (LRA), an ultrasonic acoustic output transducer, a haptic transducer for a vibrotactile system, a speaker, or other type of electromechanical actuator. In FIG. 1, the electromechanical actuator 100 is depicted as having a mechanical portion (also referred to as a tank) and a coil portion. An embodiment is described in which, in a piecewise manner, the parameters of the coil can be estimated first, and then the parameters of the tank can be subsequently estimated using the estimated parameters of the tank. The electromechanical actuator parameters can generally be referred to as Thiele / Small (TS) parameters. The electromechanical actuator 100 can be included for use in a device such as a mobile phone or other device.

[0015] The coil of the electromechanical actuator 100 is characterized as having a resistive parameter (Re), also referred to as DC resistance, in series with an inductive parameter (Le), also referred to as electrical coil inductance, which are electrical parameters of the electromechanical actuator 100. The resulting impedance of Re and Le in series is referred to as Zcoil in FIG. 1. The tank of the electromechanical actuator 100 may be modeled as a spring system with a moving mass. The tank is characterized as having mechanical parameters of the electromechanical actuator 100 in parallel with a resistance parameter at resonance (Res), an inductance parameter (Lces), which represents the compliance of the spring system, and a capacitance parameter (Cmes), which represents the moving mass of the spring system. The mechanical parameters of the electromechanical actuator 100 may also be referred to as resonance parameters. The resulting impedance of Res, Lces, and Cmes in parallel is referred to as Zmech in FIG. 1. Zmech can be equally expressed in terms of the mechanical parameters Res, resonant frequency (F0), and quality factor (Q). Res, Lces, and Cmes can be understood as the electrical analog parameters of the mechanical parameters Res, F0, and Q, which are related by equations (1)-(3).

number

[0016] 1, an excitation voltage V(t) is applied to the electromechanical actuator 100. The application of the excitation voltage V(t) induces a current I(t) that passes through Re and Le (Zcoil) in series and then through Res, Lces, and Cmes (Zmech) in parallel. More specifically, as shown, the current I(t) passes through impedance Zcoil, generating a voltage Vcoil across the coil, and the current I(t) passes through impedance Zmech, generating a voltage Vmech across the tank. The voltage Vmech may also be referred to as the back emf voltage Vbemf.

[0017] As explained in more detail below, the excitation voltage V(t) includes two components: a high-frequency (HF) tone and a low-frequency (LF) wideband signal designed to derive individual electrical and mechanical responses from the electromechanical actuator 100, which can be used as individual bases for estimating the individual electrical and mechanical parameters of the electromechanical actuator 100, in that order. That is, in a piecewise manner, the electrical parameters are estimated first and then used to estimate the mechanical parameters. As explained below, the HF and LF wideband components may be applied to the electromechanical actuator 100 in parallel, or they may be applied temporally separately in either order. While FIG. 1 illustrates applying a voltage to the electromechanical actuator 100 that induces a measured current, both of which are used to estimate the parameters of the electromechanical actuator 100, other embodiments are also contemplated in which a current is applied to the electromechanical actuator 100 that generates a measured voltage, both of which are used to estimate the parameters of the electromechanical actuator 100. A system and method for performing piecewise estimation of electrical and mechanical parameters of an electromechanical actuator 100 will now be generally described with reference to FIG.

[0018] 2 is an exemplary diagram illustrating a process for performing piecewise estimation of parameters of an electromechanical actuator, such as the electromechanical actuator 100 of FIG. 1 , according to an embodiment of the present disclosure. Generally, the electrical parameters of the coil are first estimated in time, as shown in the left portion of FIG. 2 , and the estimated electrical parameters are then used to estimate the mechanical parameters of the tank, as shown in the right portion of FIG. 2 . An excitation voltage V(t) is applied to the electromechanical actuator, which responds in the form of a measurable current I(t). More specifically, the excitation voltage V(t) applied to excite an electrical response from the coil is a short HF tone, and the excitation voltage V(t) applied to excite a resonant response (wideband response) from the tank is a short LF wideband signal (also referred to as a resonant excitation signal), each of which is described in more detail below.

[0019] The coil current I(t) response to an HF tone excitation signal V(t) is measured and used as a basis for characterizing the coil according to a first step, in addition to the excitation signal V(t). That is, the coil's electrical parameters Re and Le are estimated (e.g., using least squares estimation) based on the following equation (4), as described in more detail below with respect to FIGS. 3 and 4, where j is the imaginary square root of −1, ω is the angular frequency of the HF tone, the angular frequency value being much greater than 2π×F0, and V(t) and I(t) are complex-valued samples of the excitation signal and response, respectively.

number

[0020] The tank current I(t) response to the LF wideband excitation signal V(t) is measured and used in addition to the excitation voltage V(t) and the estimated Re and Le values ​​to then calculate the back emf voltage Vbemf (i.e., Vmech) according to equation (5).

number

number

[0021] Once the electrical and mechanical parameters have been estimated, a model of the electromechanical actuator can be constructed and the electromechanical actuator can be calibrated. Knowledge of the electromechanical actuator parameters can be used by devices that include the electromechanical actuator to adapt and optimize signals played back to the electromechanical actuator.

[0022] The piecewise nature of parameter estimation, in which separate excitation signals are targeted at the coil and tank, advantageously allows the HF excitation signal and the LF broadband excitation signal to be relatively short in duration, which may facilitate faster characterization and calibration of electromechanical actuators than previous methods. A shorter characterization time may be advantageous during the manufacture of a device (e.g., a mobile phone) incorporating an electromechanical actuator (e.g., a tactile sensation transducer of a vibrotactile sensation system), particularly when a large number of devices are being manufactured. Additionally, a short characterization time may make the regeneration of the excitation signal virtually imperceptible, thus advantageously allowing the electromechanical actuator to be characterized and calibrated on demand, for example, during production time, as well as during operation of the device by a consumer of the device. The ability to characterize and calibrate on demand may be particularly useful because the parameters of an electromechanical actuator may vary over time after the device is manufactured, for example, based on usage, temperature, aging, and other factors. More advantageously, a piecewise manner of first estimating two parameters (Re and Le) and then subsequently estimating the other three parameters (Re, Lces, and Cmes) may be more stable than estimating five parameters simultaneously. From another perspective, a piecewise manner of first estimating electrical parameters and then subsequently estimating mechanical parameters may reduce the complexity of the problem of characterizing electromechanical actuators.

[0023] Although embodiments first estimate electrical parameters and then use the estimated electrical parameters to estimate machine parameters, the HF excitation signal and the LF wideband signal may be applied in either order. Furthermore, in some embodiments, the HF excitation signal and the LF wideband excitation signal may be applied in parallel, as long as there is sufficient spacing between the HF and LF wideband signals so that they do not produce mutually interfering harmonics. In other words, the calculation / estimation of parameters in a piecewise manner is a distinctly different operation from the application of the HF and LF wideband excitation signals. More specifically, while the electrical parameters are estimated first and the estimated electrical parameters are then used to estimate machine parameters, the HF excitation signal need not be applied temporally before the application of the LF wideband excitation signal, but instead may be applied in parallel with or after it.

[0024] FIG. 3 is an exemplary diagram illustrating a first step of a piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure. FIG. 3 includes a circuit diagram 300 illustrating the application of an HF excitation signal V(t) to the coil of the electromechanical actuator 100 of FIG. 1 to derive a response in the form of a current I(t), an estimator 301, and example waveforms of V(t) and I(t). In one embodiment, the HF excitation signal V(t) is a pilot tone burst at 2 kHz, although other embodiments are contemplated having tone bursts at other frequencies that are sufficiently separated from the frequency band of the LF broadband excitation signal used in the second step of the piecewise characterization process, described in more detail below. In the example of FIG. 3 , approximately 240 data samples of V(t) and I(t) were collected at 48 kHz over approximately 5 milliseconds, which has been empirically determined to be sufficient for a tested sample of electromechanical actuator to produce accurate results. A least squares estimator 301 estimates Re and Le using the HF excitation signal V(t) and the response current I(t).

[0025] The least squares estimator 301 includes a regressor 302 and an output function 308 that outputs a voltage V(t). The output function 308 includes an offset 304 and a linear function block 306. The offset 304 may correspond to the offset of a current monitor that measures the current I(t) and requires mitigation (e.g., using a sense resistor of known value and an analog-to-digital converter that converts the measured voltage across the resistor) to more accurately estimate Re and Le. The regressor 302 receives as input the current I(t) and the fed-back voltage V(t). The output of the regressor 302 is provided as an input to the linear function block 306. The offset 304 is subtracted from the output of the linear function block 306 to produce the voltage V(t). In one embodiment, the estimator 301 uses well-known least squares estimation methods to estimate Re and Le using the voltage V(t) and current I(t) according to equation (4) above using the collected data samples. Although an embodiment is described that employs least squares estimation to estimate Re and Le, they may also be estimated by other well-known estimation methods, including, but not limited to, least mean squares (LMS) estimation or other iterative methods, as well as adaptive filtering. Additionally, the HF excitation signal V(t) includes a minimum offset to facilitate accurate estimation of Re and Le.

[0026] As will be more clearly understood from the description of the second step of piecewise characterization below, accurate estimation of mechanical parameters by the second step may require accurate estimation of electrical parameters by the first step. This contrasts with previous methods for identifying mechanical impedance, such as those described in U.S. Pat. Nos. 10,726,683 and 11,263,877 (each of which is incorporated by reference herein in its entirety for all purposes). In these patents, back-emf voltages are estimated at two different tone frequencies, and the difference (magnitude and phase) of the back-emf voltages is calculated. In such an approach, "the estimates of DC resistance Re and inductance Le may not need to be accurate (e.g., an error within about 10% may be acceptable); therefore, fixed values ​​from offline calibration or from datasheet specifications may be sufficient." The tolerance of Re and Le errors in the patented method results from the fact that errors in the voltage estimates caused by errors in Re and Le will essentially be eliminated by taking the difference. In contrast, as described below, the second step of an embodiment of the present disclosure uses the estimated back-emf voltage (Vbemf) directly (i.e., by not differencing) to estimate the machine parameters. Therefore, mitigating the current monitor offset can significantly improve the accuracy of the characterization of the electromechanical actuator. In addition, compensation for the shift in the real part of Zcoil, described below with respect to FIG. 4, can also be employed to improve the accuracy of the characterization of the electromechanical actuator.

[0027] FIG. 4 is an example diagram illustrating high-frequency compensation of low-frequency estimation employed in the first step of the piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure. FIG. 4 includes a graph illustrating the measured variation of the DC coil resistance Re (referred to as Re in FIG. 4 ) of the coil impedance Z of an example instance of the electromechanical actuator 100 at different frequencies. In theory, the value of the DC coil resistance Re should be fixed and not change with frequency, as shown by the line labeled Re in FIG. 4 . However, through measurements of the imaginary and real parts of the coil impedance (shown as curves Z and Z, respectively, in FIG. 4 ), it has been observed that the real part Z shifts higher in value as frequency increases. Parasitic capacitance within the coil (shown as Z in FIG. 4 ) is hypothesized to account for the shift in Z, and the skin effect may also contribute to the shift.

[0028] In one embodiment, the first step compensates for the Zreal shift with a fixed scaling factor. In one embodiment, for each of many samples of the electromechanical actuator, a high-frequency estimate of ReDC was taken when the coil was slightly heated. Also, for each sample, a low-frequency estimate of ReDC was obtained by playing a pilot tone at a very low frequency. A scaling factor was obtained as the difference between the high-frequency and low-frequency estimates. That is, the scaling factor is a pre-determined value by which the high-frequency estimate of Re is multiplied to obtain a low-frequency estimate of Re. The application of the scaling factor to the value of Re determined by estimator 301 of FIG. 3 may be referred to herein as high-frequency compensation of the low-frequency estimate. Because estimation of the mechanical parameters of the electromechanical actuator, following the second piecewise step, may be sensitive to the accuracy of the Re estimate, high-frequency compensation of the low-frequency estimate may improve the accuracy of the mechanical parameter estimate.

[0029] FIG. 5 is an exemplary graph of the magnitude of the impedance Z of an electromechanical actuator's tank as a function of frequency, according to an embodiment of the present disclosure. As shown in FIG. 5, the center frequency at which the tank impedance is greatest is the tank's resonant frequency F, R is the peak impedance value, and the bandwidth around the peak is related to the quality factor Q. In the example of FIG. 5, F is approximately 200 Hz, Q is approximately 8, and R is approximately 1.6 ohms, although different electromechanical actuators may have different values ​​of R, F, and Q. The LF wideband excitation signal, an example of which is shown in FIG. 6, is designed to have a bandwidth wide enough to encompass the tank impedance peaks to enable estimation of the tank's mechanical parameters.

[0030] FIG. 6 includes separate time and frequency domain graphs of an exemplary LF broadband excitation signal employed in the second step of the piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure. In the exemplary embodiment of FIG. 6, the LF broadband excitation signal includes two cycles of a sinusoidal waveform multiplied by a window. An integer number of cycles is used in the LF broadband excitation signal to avoid introducing any DC or HF content into the excitation waveform. Although two cycles of a sinusoidal waveform are shown in FIG. 6, other integer numbers of cycles may also be employed. The window has a smoothing effect by gradually reducing the data values ​​to zero at the end of the sinusoidal waveform, as can be observed in FIG. 6 (and in the example of FIG. 8). Examples of window types are shown in FIG. 6 and may include, but are not limited to, a rectangular window, a Nuttall window, a flat-top window, and a Gaussian window. In certain embodiments, a Nuttall window provides a smooth waveform with a sufficiently wide bandwidth, as can be observed from the frequency domain graph of FIG. 6. The transient on / off switching of the LF broadband excitation signal provides a bandwidth wide enough to cover the tank impedance peaks to allow estimation of the tank's mechanical parameters.

[0031] FIG. 7 is an exemplary diagram illustrating a second step of a piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure. FIG. 7 includes an estimator 701 and a circuit diagram 700 illustrating the application of an LF broadband excitation signal to the tank of the electromechanical actuator 100 of FIG. 1 to derive a response in the form of a current I(t) and a back-emf voltage Vbemf, for example, as described with respect to FIG. 6. As described above with respect to FIG. 6, the LF broadband excitation signal is designed to excite a band around the resonant frequency of the tank, i.e., around F0, and wide enough to cover the resonant peaks shown in FIG. 5, for example. The response current I(t) is measured and used to calculate the back-emf voltage Vbemf(t) according to equation (5) above, in addition to the R and L estimated from the first step. The measured current I(t) and the calculated Vbemf(t) are then used by estimator 701 to estimate the mechanical parameters Res, Lces, and Cmes (or alternatively Res, F0, and Q) of the tank of the electromechanical actuator.

[0032] Least squares estimator 701 includes a regressor 702 and an output function 708 that outputs the back-emf voltage Vbemf(t). Output function 708 includes an offset 704 (e.g., of a current monitor as described above that requires mitigation to more accurately estimate Res, Lces, and Cmes) and a linear function block 706. Regressor 702 receives as input the current I(t) and the fed-back back-emf voltage Vbemf(t). The output of regressor 702 is provided as an input to linear function block 706. Offset 704 is subtracted from the output of linear function block 706 to produce the back-emf voltage Vbemf(t). In one embodiment, the estimator 701 uses least squares estimation to estimate Res, Lces, and Cmes using the collected data samples according to equation (6) above using the back emf voltage Vbemf(t) and current I(t), although other estimation methods may also be employed, as described above with respect to FIG. 3.

[0033] 8 includes exemplary graphs of HF tone and LF broadband excitation signals employed in a piecewise electromechanical actuator characterization process according to an embodiment of the present disclosure. In the exemplary embodiment of FIG. 8, the HF tone portion of the excitation signal includes 10 cycles of a 2 kHz sinusoidal pilot tone waveform having a burst duration of approximately 5 milliseconds, designed to excite the coil for use in estimating Re and Le, as described above, and the LF broadband portion of the excitation signal includes two cycles of a 200 Hz sinusoidal waveform multiplied by a Nuttall window, designed to excite the tank for use in estimating Re, F, and Q (or alternatively, Re, L, and C), as described above.

[0034] In the example of Figure 8, the regeneration of the HF excitation signal is performed first in time, followed by the regeneration of the LF broadband excitation signal. However, in other embodiments, the regeneration of the LF broadband excitation signal is performed first in time, followed by the regeneration of the HF excitation signal. However, in both embodiments, the electrical parameters (Re, Le) are estimated first and then used to estimate the mechanical parameters (Re, L, C or Res, F, and Q). Furthermore, in other embodiments, the regeneration of the HF excitation signal and the regeneration of the LF broadband excitation signal are performed in parallel, as long as the HF excitation signal and the LF broadband excitation signal do not produce harmonics that interfere with each other.

[0035] In the example of FIG. 8 , the peak amplitude of the HF excitation signal is approximately 0.2 volts. However, other embodiments are contemplated in which the peak amplitude of the HF excitation signal is other values. Furthermore, the HF excitation signal may be reduced in amplitude (e.g., to reduce or avoid disruption to the user) if more cycles are played so that a similar amount of energy is applied to excite the electromechanical actuator. In the example of FIG. 8 , the peak amplitude of the LF wideband excitation signal is approximately 1 volt; however, embodiments with other peak amplitudes are contemplated. Furthermore, the LF wideband excitation signal may be reduced in amplitude if multiple wideband cycles are played, thereby improving the signal-to-noise ratio (SNR). Still further, playing the HF excitation signal and / or the LF wideband excitation signal and measuring the response thereto may be repeated multiple times to improve the SNR.

[0036] In the example of FIG. 8 , the HF excitation signal is a 2 kHz sinusoidal tone. However, other embodiments are contemplated in which a high frequency other than 2 kHz is employed. For example, the HF excitation signal may be substantially higher than the resonant frequency of the electromechanical actuator, e.g., high enough to avoid overlap of the electromechanical actuator's frequency responses to the separate HF and LF broadband excitation signals. From another perspective, given the electrical (coil) and mechanical (tank) configuration of the electromechanical actuator, the frequency of the HF excitation tone may be high enough to avoid interference with the coil's response to the HF excitation signal from the electromechanical actuator's mechanical resonance. In other words, the HF excitation signal may be outside the resonant frequency band of the electromechanical actuator, such as the resonant frequency band shown in FIG. 5 . In some embodiments, the HF excitation signal tone frequency may be within a frequency range centered about 5 to 10 times the mechanical resonant frequency of the electromechanical actuator to create a significant gap in frequency between the HF excitation signal and the LF broadband excitation signal.

[0037] In the example of FIG. 8 , the LF broadband excitation signal includes two cycles of a 200 Hz sinusoidal waveform multiplied by a Nuttall window. While the resonant frequency F0 may be one of the parameters of the electromechanical actuator being determined by the described embodiments, a range of possible resonant frequencies may be experimentally predetermined by testing a sample of instances of the electromechanical actuator and used to design the LF broadband excitation signal. The range of possible resonant frequencies may then be used to select frequencies (multiplied by the window) for the sinusoidal waveform of the LF broadband excitation signal centered within that range, and the window may be selected to spectrally cover the band of the range of resonant frequencies. In other words, the predetermined range may be understood as an a priori guess at the resonant frequency, which is more accurately estimated using the parameter estimation embodiments of the present disclosure. Thus, although the frequency of the sinusoidal waveform in the example of FIG. 8 is 200 Hz, other frequencies may also be selected based on the range of possible pre-determined resonant frequencies. Furthermore, as explained above, the sinusoidal waveform may be multiplied by other types of windows.

[0038] In an embodiment, an HF excitation signal and / or an LF wideband excitation signal may be applied to the electromechanical actuator and its response thereto measured multiple times to improve SNR. Additionally, between the multiple times that the LF wideband excitation signal is applied, one or more of the parameters of the LF wideband excitation signal, such as the frequency of the sinusoidal waveform, the amplitude of the sinusoidal waveform, the integer number of cycles of the sinusoidal waveform, and the type of window by which the sinusoidal waveform is multiplied, may be adjusted.

[0039] Provided that signal processing calculation times can be small compared to excitation and measurement times, it can be observed that only about tens of milliseconds can be required to characterize an electromechanical actuator in the described piecewise manner, which is a significant time reduction over previous methods and can provide the advantages described herein. Characterization time can vary depending on various factors such as the resonant frequency of the electromechanical actuator, the number of samples in each of the HF and LF excitation signal components, and the number of times the excitation signal is applied and the response is measured.

[0040] Still further, embodiments of the present disclosure may benefit from lower power consumption compared to conventional methods, such as the chirp method described above. Conventional methods employing chirp stimulation may consume relatively large amounts of power due to the long duration required to sweep through a frequency range to elicit a response from both the coil and the tank. In contrast, embodiments of the present disclosure decompose the stimulation into two parts separated by a gap: specifically, HF and LF components directed to the coil and the tank, respectively, each of which may be relatively narrow bands (in practice, the HF stimulation may be a tone), each of which may be a few cycles, and thus may consume less power compared to conventional chirp stimulation methods. For similar reasons, embodiments of the present disclosure may consume less power than conventional impulse methods due to the relatively wider amplitude of the required impulse.

[0041] In particular, those skilled in the art having the benefit of this disclosure will understand that the various operations described herein, particularly in connection with the Figures, may be implemented by other circuitry or other hardware components. The order in which the operations of a given method are performed may be changed unless otherwise indicated, and various elements of the systems illustrated herein may be added, reordered, combined, omitted, modified, etc. It is intended that the present disclosure encompass all such modifications and variations, and therefore the foregoing description should be considered in an illustrative and not a restrictive sense.

[0042] Similarly, although the present disclosure refers to specific embodiments, certain modifications and variations can be made to those embodiments without departing from the scope and breadth of the present disclosure. Additionally, any benefits, advantages, or solutions to problems described herein with respect to specific embodiments are not intended to be construed as critical, required, or essential features or elements.

[0043] Further embodiments will similarly be apparent to those skilled in the art with the benefit of this disclosure, and such embodiments should be considered to be encompassed herein. All examples and conditional language recited herein are intended for educational purposes to aid the reader in understanding the disclosure and concepts contributed by the inventors to further the art, and are to be construed as without limitation to such specifically recited examples and conditions.

[0044] This disclosure encompasses all changes, substitutions, variations, alterations, and modifications to the exemplary embodiments herein that would be understood by one of ordinary skill in the art. Similarly, where appropriate, the appended claims also encompass all changes, substitutions, variations, alterations, and modifications to the exemplary embodiments herein that would be understood by one of ordinary skill in the art. Also, a reference in the appended claims to an apparatus or system, or a component of an apparatus or system, being adapted, arranged, capable, configured, enabled, operative, or operating to perform a particular function encompasses that apparatus, system, or component, so long as it is so adapted, arranged, capable, configured, enabled, operative, or operating, regardless of whether itself or that particular function is activated, turned on, or unlocked.

[0045] Finally, software can effect or configure the functions, processes, and / or instructions of the apparatus and methods described herein. This can be accomplished using common programming languages ​​(e.g., C, C++), hardware description languages ​​(HDLs), including Verilog HDL, VHDL, etc., or other available programs. Such software can be located in any known non-transitory computer-readable medium, such as magnetic tape, semiconductor, magnetic or optical disk (e.g., CD-ROM, DVD-ROM, etc.), network, wireline, or another communications medium, having instructions stored thereon that are capable of effecting or configuring the apparatus and methods described herein.

[0046] To aid the Patent Office and any reader of this application and any patents issued thereon in interpreting the claims appended hereto, Applicants wish to indicate that none of the appended claims or claim elements are intended to invoke 35 U.S.C. Section 112(f) unless the words "means for" or "step for" are expressly used within a particular claim. Furthermore, use of the term "configured to" is not intended to invoke 35 U.S.C. Section 112(f).

Claims

1. 1. A method comprising: applying a high frequency signal to an electromechanical actuator and measuring a first response of the electromechanical actuator to the high frequency signal; estimating electrical parameters of the electromechanical actuator based on the first response; applying a low frequency broadband signal to the electromechanical actuator and measuring a second response of the electromechanical actuator to the low frequency broadband signal; estimating mechanical parameters of the electromechanical actuator based on the second response and the estimated electrical parameters; and A method comprising:

2. The method of claim 1 , wherein applying the high frequency signal and applying the low frequency wideband signal are performed in parallel.

3. The method of claim 2 , wherein the high frequency signal and the low frequency wideband signal are selected so that they do not produce mutually interfering harmonics.

4. The method of claim 1 , wherein applying the high frequency signal is performed prior to applying the low frequency wideband signal.

5. The method of claim 1 , wherein applying the high frequency signal is performed after applying the low frequency wideband signal.

6. The method of claim 1 , further comprising estimating the electrical and mechanical parameters of the electromechanical actuator during calibration of the electromechanical actuator during manufacture of a device including the electromechanical actuator.

7. The method of claim 1 , further comprising estimating the electrical and mechanical parameters of the electromechanical actuator during operation by a consumer of a device including the electromechanical actuator.

8. The method of claim 1 , wherein the electrical parameters and the mechanical parameters are acquired in less than 50 milliseconds.

9. 10. The method of claim 1, wherein applying the high frequency signal and measuring the first response and / or applying the low frequency broadband signal and measuring the second response is repeated multiple times to improve signal to noise ratio.

10. 2. The method of claim 1, wherein the low-frequency broadband signal spectrally covers a frequency band centered around a range of mechanical resonant frequencies that are experimentally predetermined from a sample of instances of the electromechanical actuator.

11. The method of claim 1 , wherein the low-frequency wideband signal comprises a sinusoidal waveform multiplied by a window.

12. applying the low frequency broadband signal and measuring the second response is repeated a plurality of times; For each of said plurality of times, the frequency of the sinusoidal waveform; the amplitude of the sinusoidal waveform; an integer number of cycles of said sinusoidal waveform; The window type; The method of claim 11 , wherein one or more of:

13. The method of claim 11 , wherein the high frequency signal is sufficiently higher than the frequency of the sinusoidal waveforms of the low frequency wideband signal to avoid overlap within their individual frequency responses.

14. The method of claim 1 , wherein the high frequency signal is sufficiently high to avoid interference with the first response from mechanical resonances of the electromechanical actuator.

15. The method of claim 14 , wherein the high frequency signal is about an order of magnitude higher than the resonant frequency of the electromechanical actuator.

16. The method of claim 1 , wherein the high frequency signal is outside a band of resonant frequencies of the electromechanical actuator.

17. estimating the machine parameters calculating a back emf voltage using the estimated electrical parameters and the measured second response; estimating the machine parameters using the calculated back emf voltage and the measured second response; and The method of claim 1 , comprising:

18. The electrical parameters include direct current (DC) electrical resistance (Re); estimating the electrical parameters estimating Re based on the first response; and applying a predetermined scaling factor to the estimated Re to compensate for shifts in the real component of the impedance of the coil portion of the electromechanical actuator at high frequencies; The method of claim 1 , comprising:

19. The method of claim 1 , wherein estimating the electrical parameter includes compensating for an offset in a circuit used to measure the first response.

20. the electrical parameters include a direct current (DC) electrical resistance (Re) and an electrical coil inductance (Le) of the electromechanical actuator; The method of claim 1 , wherein the mechanical parameters include a resistance at resonance (Res), a resonant frequency (F0), and a quality factor (Q) of the electromechanical actuator, or equivalents thereof.

21. 1. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing a method, the method comprising: applying a high frequency signal to an electromechanical actuator and measuring a first response of the electromechanical actuator to the high frequency signal; estimating electrical parameters of the electromechanical actuator based on the first response; applying a low frequency broadband signal to the electromechanical actuator and measuring a second response of the electromechanical actuator to the low frequency broadband signal; estimating mechanical parameters of the electromechanical actuator based on the second response and the estimated electrical parameters; and 1. A non-transitory computer-readable storage medium comprising:

22. 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21, wherein the applying the high-frequency signal and the applying the low-frequency wideband signal are performed in parallel.

23. the high frequency signal and the low frequency broadband signal are selected so that they do not produce mutually interfering harmonics; 23. A non-transitory computer readable storage medium having computer program instructions stored thereon for implementing the method of claim 22.

24. applying the high frequency signal is performed prior to applying the low frequency broadband signal.

22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

25. applying the high frequency signal is performed after applying the low frequency broadband signal.

22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

26. further comprising estimating the electrical and mechanical parameters of the electromechanical actuator during calibration of the electromechanical actuator during manufacture of a device including the electromechanical actuator.

22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

27. and estimating the electrical and mechanical parameters of the electromechanical actuator during operation by a consumer of a device including the electromechanical actuator.

22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

28. The electrical parameters and the mechanical parameters are acquired in less than 50 milliseconds.

22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

29. applying the high frequency signal and measuring the first response and / or applying the low frequency broadband signal and measuring the second response are repeated multiple times to improve the signal to noise ratio; 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

30. the low frequency broadband signal spectrally encompasses a frequency band centered around a range of mechanical resonant frequencies, experimentally predetermined from a sample of instances of the electromechanical actuator; 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

31. the low frequency wideband signal comprises a sinusoidal waveform multiplied by a window; 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

32. applying the low frequency broadband signal and measuring the second response is repeated a plurality of times; For each of said plurality of times, the frequency of the sinusoidal waveform; the amplitude of the sinusoidal waveform; an integer number of cycles of said sinusoidal waveform; The window type; one or more of the following are adjusted:

32. A non-transitory computer readable storage medium having computer program instructions stored thereon for implementing the method of claim 31.

33. the high frequency signal being sufficiently higher than the frequency of the sinusoidal waveform of the low frequency wideband signal to avoid overlap within their individual frequency responses; 32. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 31.

34. the high frequency signal is sufficiently high to avoid interference with the first response from mechanical resonances of the electromechanical actuator; 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

35. the high frequency signal is approximately one order of magnitude higher than the resonant frequency of the electromechanical actuator; 35. A non-transitory computer readable storage medium having computer program instructions stored thereon for implementing the method of claim 34.

36. the high frequency signal is outside the band of resonant frequencies of the electromechanical actuator; 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

37. estimating the machine parameters calculating a back emf voltage using the estimated electrical parameters and the measured second response; estimating the machine parameters using the calculated back emf voltage and the measured second response; and Including, 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

38. The electrical parameters include direct current (DC) electrical resistance (Re); estimating the electrical parameters estimating Re based on the first response; and applying a predetermined scaling factor to the estimated Re to compensate for shifts in the real component of the impedance of the coil portion of the electromechanical actuator at high frequencies; Including, 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

39. estimating the electrical parameter includes compensating for an offset of a circuit used to measure the first response.

22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

40. the electrical parameters include a direct current (DC) electrical resistance (Re) and an electrical coil inductance (Le) of the electromechanical actuator; the mechanical parameters include the resistance at resonance (Res), the resonant frequency (F0), and the quality factor (Q) of the electromechanical actuator, or equivalents thereof; 22. A non-transitory computer-readable storage medium having computer program instructions stored thereon for implementing the method of claim 21.

Images