A Simulation Test Method for Sound Pressure Distribution of a Piezoelectric MEMS Loudspeaker

By matching the measured electrical signals with the preset structural transmission coefficients and iteratively adjusting the acoustic impedance boundary using a simulation model, the problem of distinguishing between airflow leakage and structural loss in piezoelectric MEMS loudspeakers was solved, achieving accurate simulation and optimization of low-frequency sound pressure distribution.

CN122698920BActive Publication Date: 2026-11-10HUNAN HUICHENG TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611199971.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-08-10
Publication Date
2026-11-10
Estimated Expiration
2046-08-10

AI Technical Summary

Technical Problem

Existing experimental testing methods are insufficient to accurately distinguish between airflow leakage and structural losses in piezoelectric MEMS loudspeakers. This results in simulation models being unable to provide accurate flow resistance boundary calibration and failing to effectively address nonlinear coupling issues under high voltage drive, especially in the low-frequency, high-amplitude range where sound pressure distribution prediction is inaccurate.

Method used

By matching the measured electrical signals with the preset structural transmission coefficients, the characteristic value of airflow leakage is determined. The acoustic impedance boundary is then iteratively adjusted in conjunction with the simulation model until the characteristic deviation meets the convergence condition, thereby achieving accurate simulation of sound pressure distribution.

Benefits of technology

It significantly improves the prediction accuracy of low-frequency sound pressure distribution under high voltage conditions, solves the technical pain point that conventional power testing cannot distinguish between structural losses and airflow losses, and greatly improves the consistency between simulation prediction curves and measured data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122698920B_ABST
    Figure CN122698920B_ABST
Patent Text Reader

Abstract

This application relates to a simulation test method for sound pressure distribution of a piezoelectric MEMS loudspeaker. The method includes: S11, determining the measured airflow leakage characteristic value based on the matching degree between the energy distribution of the measured electrical signal of the device under test and the preset structural transfer coefficient, and calculating the acoustic impedance boundary based on the measured slit opening value of the device under test; S12, running a preset simulation model to obtain a virtual waveform under the acoustic impedance boundary, and extracting the virtual airflow leakage characteristic value based on the virtual waveform; S13, calculating the characteristic deviation between the measured airflow leakage characteristic value and the virtual airflow leakage characteristic value, and adjusting the acoustic impedance boundary in the simulation model according to the direction and value of the characteristic deviation; S14, repeating steps S12 and S13 until the characteristic deviation meets the preset convergence condition, and outputting the sound pressure distribution data under the acoustic impedance boundary. This application can improve the prediction accuracy of low-frequency sound pressure distribution.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of simulation and testing technology, and in particular to a method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker. Background Technology

[0002] Piezoelectric MEMS loudspeakers utilize the inverse piezoelectric effect to drive ultrathin cantilever beams or diaphragms to produce sound, making them a mainstream solution for the miniaturization of portable audio devices. To release residual stress generated during microfabrication, the diaphragm edge is typically designed with micrometer-wide stress-relieving slits. In simulation prediction, the industry standard practice is to establish a computational model based on an ideal fixed boundary or a constant damping boundary. However, under actual high-voltage driving conditions, piezoelectric MEMS loudspeakers face complex nonlinear coupling problems: the hysteresis properties of piezoelectric materials cause structural vibrations to generate harmonics of various orders, and large-amplitude vibrations cause significant periodic opening and closing of the slits at the diaphragm edge. Since air resistance at the microscale is extremely sensitive to the gap size, this dynamic opening and closing induces nonlinear air leakage, meaning that the acoustic impedance boundary changes drastically with the driving voltage and instantaneous displacement. Existing experimental testing methods (such as laser vibration measurement or electrical impedance analysis) can only obtain total power loss or single displacement data. It is difficult for testers to separate the individual airflow leakage component from the total loss, resulting in the inability to provide accurate flow resistance boundary calibration basis for simulation models. Summary of the Invention

[0003] To address the aforementioned technical problems, the purpose of this application is to provide a simulation and testing method for the sound pressure distribution of a piezoelectric MEMS loudspeaker. The specific technical solution adopted is as follows: In a first aspect, a method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker is provided, the method comprising: S11. Based on the matching degree between the energy distribution of the measured electrical signal of the device under test and the preset structural transmission coefficient, determine the measured air leakage characteristic value, and calculate the acoustic impedance boundary based on the measured slit opening value of the device under test. S12. Run the preset simulation model to obtain the virtual waveform under the acoustic impedance boundary, and extract the virtual airflow leakage characteristic value based on the virtual waveform. S13. Calculate the characteristic deviation between the measured airflow leakage characteristic value and the virtual airflow leakage characteristic value, and adjust the acoustic impedance boundary in the simulation model according to the direction and value of the characteristic deviation. S14. Repeat steps S12 and S13 until the characteristic deviation meets the preset convergence condition, and output the sound pressure distribution data under the acoustic impedance boundary.

[0004] Optionally, the measured airflow leakage characteristic value is determined based on the matching degree between the energy distribution of the measured electrical signal of the device under test and the preset structural transfer coefficient, including: Frequency domain analysis was performed on the measured electrical signals of the device under test to obtain the relative harmonic amplitude sequences of the inner electrode and the outer electrode. Based on the preset structural transmission coefficient, the relative harmonic amplitude sequence of the inner electrode and the relative harmonic amplitude sequence of the outer electrode are weighted and matched to obtain the structural vibration consistency coefficient. The measured characteristic value of airflow leakage is determined based on the structural vibration consistency coefficient.

[0005] Optionally, the measured characteristic value of airflow leakage is determined based on the structural vibration consistency coefficient, including: The measured airflow leakage characteristic value is determined based on the difference between the preset value and the structural vibration consistency coefficient.

[0006] Optionally, the acoustic impedance boundary is calculated based on the measured slit opening value of the device under test, including: Obtain the measured slit opening value of the device under test under the target driving voltage; Obtain the dynamic viscosity constant of air, the width of the slit, the length of the slit, and the initial gap of the slit; The acoustic impedance boundary is determined based on the sum of the measured slit opening value and the initial slit gap, the product of the dynamic viscosity constant of air and the length of the slit, and the width of the slit.

[0007] Optionally, a preset simulation model is run to obtain virtual waveforms under acoustic impedance boundaries, including: Run the preset simulation model and perform transient calculations under the current acoustic impedance boundary to obtain the virtual voltage and virtual current waveforms at the inner electrode port and the outer electrode port; wherein, the virtual waveforms include the virtual voltage and virtual current waveforms at the inner electrode port and the outer electrode port.

[0008] Optionally, virtual airflow leakage feature values ​​are extracted based on the virtual waveform, including: Frequency domain analysis is performed on the virtual waveform to generate virtual relative harmonic amplitude sequences for the inner electrode and the outer electrode, respectively. Based on the preset structural transmission coefficient, the virtual relative harmonic amplitude sequence of the inner electrode and the virtual relative harmonic amplitude sequence of the outer electrode are weighted and matched to obtain the virtual structural vibration consistency coefficient. The virtual airflow leakage characteristic value is determined based on the difference between the preset value and the vibration consistency coefficient of the virtual structure.

[0009] Optionally, the characteristic deviation between the measured airflow leakage characteristic value and the virtual airflow leakage characteristic value is calculated, including: The difference between the virtual airflow leakage characteristic value and the measured airflow leakage characteristic value is calculated to obtain the characteristic deviation.

[0010] Optionally, the acoustic impedance boundary in the simulation model can be adjusted according to the direction and value of the characteristic deviation, including: The direction of adjustment of the acoustic impedance boundary is determined by the sign of the characteristic deviation. The adjustment range is determined by the ratio of the absolute value of the characteristic deviation to the measured characteristic value of airflow leakage.

[0011] Optionally, the adjustment direction of the acoustic impedance boundary is determined based on the sign of the characteristic deviation, including: When the characteristic deviation is positive, increase the acoustic impedance boundary; When the characteristic deviation is negative, reduce the acoustic impedance boundary.

[0012] Optionally, the preset convergence conditions include: the ratio of the absolute value of the characteristic deviation to the measured airflow leakage characteristic value is less than a preset convergence threshold.

[0013] This application offers the following advantages: It determines the measured airflow leakage characteristic value based on the matching degree between the energy distribution of the measured electrical signal of the device under test (DUT) and the preset structural transfer coefficient, achieving precise decoupling between structural nonlinear loss and airflow leakage loss. The acoustic impedance boundary is calculated based on the measured slit opening value of the DUT. A simulation model is run to obtain virtual waveforms and extract virtual airflow leakage characteristic values. The characteristic deviation between the measured and virtual airflow leakage characteristic values ​​is calculated, and the acoustic impedance boundary is adjusted according to the direction and value of the characteristic deviation. Through repeated iterations until the characteristic deviation meets the convergence condition, the acoustic impedance boundary of the simulation model adaptively approximates the real dynamic leakage environment. Through the dual mechanism of characteristic decoupling and closed-loop calibration, it effectively solves the problem that existing simulation methods using static flow resistance parameters cannot reflect dynamic leakage effects, significantly improving the prediction accuracy of low-frequency sound pressure distribution under high voltage conditions, and addressing the technical pain point that conventional power testing cannot distinguish between structural loss and airflow loss. By iteratively correcting the acoustic impedance boundary, the low-frequency sound pressure attenuation phenomenon caused by the dynamic opening and closing of the slit was successfully reproduced. This significantly improved the consistency between the simulation prediction curve and the measured data across the entire frequency band (especially in the low-frequency large amplitude range), thereby enhancing the prediction accuracy of the low-frequency sound pressure distribution and providing a reliable basis for optimizing the acoustic performance of the device. Attached Figure Description

[0014] Figure 1 This is a flowchart of a sound pressure distribution simulation test method for a piezoelectric MEMS loudspeaker in one embodiment; Figure 2This is a schematic diagram of the structure of a piezoelectric MEMS loudspeaker sound pressure distribution simulation test system in one embodiment. Detailed Implementation

[0015] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0016] The following, with reference to the accompanying drawings, details the specific scheme of the sound pressure distribution simulation test method for a piezoelectric MEMS loudspeaker provided in this application. For example... Figure 1 As shown, the method includes: S11. Based on the matching degree between the energy distribution of the measured electrical signal of the device under test and the preset structural transmission coefficient, determine the measured airflow leakage characteristic value, and calculate the acoustic impedance boundary based on the measured slit opening value of the device under test.

[0017] The device under test can be a piezoelectric MEMS loudspeaker, or a piezoelectric micromechanical ultrasonic transducer, a piezoelectric micromechanical microphone, or other piezoelectric micromechanical acoustic sensors. The subsequent steps of this application will be described using a piezoelectric MEMS loudspeaker as an example.

[0018] To comprehensively capture the behavior of devices under dynamic conditions, a testing system capable of simultaneously recording electrical response and mechanical displacement needs to be built. The operator first fixes the piezoelectric MEMS speaker sample to be tested on a free-field testing bracket, ensuring its acoustic radiation surface is unobstructed. The system connection is as follows: 1. Electrical Excitation and Measurement: A dual-channel power analyzer was used, with its voltage and current probes connected in parallel and series to the inner and outer electrode drive ports of the speaker, respectively. This device was used to record instantaneous voltage data. ) and instantaneous current data ( ).

[0019] 2. Mechanical Displacement Measurement: A laser Doppler vibrometer (LDV) is used, with its optical probe focused on the cantilever end surface inside the slit at the edge of the speaker diaphragm. This device is used to monitor the vertical displacement waveform of this critical location relative to the stationary base in real time to obtain the physical opening state of the slit.

[0020] The data acquisition process employs a voltage scanning strategy: the frequency of the sinusoidal excitation signal is set to the first-order resonant frequency of the loudspeaker (e.g., 1.2kHz). The amplitude of the driving voltage is set. The scan range starts from the linear operating region voltage of the device (e.g., 0.1V) and gradually increases in fixed voltage steps (e.g., 0.5V) until the maximum rated voltage of the device (e.g., 15V) is reached.

[0021] At each voltage scan point, the system synchronously triggers the acquisition and recording of the following two sets of data: 1. Electrical waveform set: time-domain waveforms of voltage and current of the inner and outer electrodes; 2. Geometric opening value: based on the displacement waveform recorded by the laser vibrometer, half of its peak-to-peak value is calculated and defined as the measured slit opening value at that voltage point. (Unit: micrometer), which intuitively reflects the physical opening of the slit.

[0022] The preset structural transfer coefficient can be obtained from either of the following two calibration methods, A and B: Method A: The operator temporarily seals the stress relief slit of the test sample by applying adhesive or covering it with tape, or places the sample in a vacuum chamber for testing. In this case, the gas leakage path is physically cut off, and the device's losses originate solely from structural damping and dielectric loss. Under these conditions, the maximum rated voltage of the test device is applied, and the electrical signals of the inner and outer electrodes are acquired and subjected to frequency spectrum analysis (FFT). The first external electrode is calculated... Second harmonic amplitude With the internal electrode Second harmonic amplitude The ratio of the two values ​​directly yields the true structural transfer coefficient. The calculation formula is: ; Structural transfer factor Characterizing the internal electrode under ideal, leak-free conditions The second harmonic energy is transferred to the external electrode. The inherent proportion of subharmonics, for subharmonics of the same order (i.e. ), structural transfer coefficient It accurately reflects the structural coupling efficiency at this order.

[0023] Method B: If physical blocking is not possible, test data from the small-signal linear region can be used for deduction. Under extremely low voltage (e.g., 0.1V) driving, the device is in the linear operating region and airflow leakage is negligible. At this time, the signal is acquired and the fundamental frequency is calculated. The amplitude ratio is used to obtain the reference transfer coefficient. Based on the homology assumption of piezoelectric structures, meaning that the generation mechanism of each order harmonic is the same, it can be approximated that the transmission ratio of higher-order harmonics is the same as that of the fundamental wave. Therefore, all transmission coefficients of the same order are set. All equal to , that is Although this method is approximate, it can effectively provide structural references within the limits of engineering accuracy.

[0024] The collected data is then processed to generate a lookup table for use by the simulation model. The electrical waveforms collected at each voltage point are input into the feature extraction algorithm described in the subsequent steps to calculate the corresponding measured airflow leakage characteristic values. Subsequently, the following two sets of data mapping relationships are constructed: Master reference relationship: Established based on drive voltage As an index, based on measured airflow leakage characteristic values The lookup table is set as the target, and this relationship will serve as the convergence target for subsequent simulation closed-loop calibration.

[0025] Auxiliary reference relationship: Establish the measured slit opening value Reference table of fluid theory impedance values. Using the parallel plate flow resistance formula (Lubrication Theory), based on the measured slit opening value... Calculate the corresponding theoretical acoustic impedance reference value. ,in This relationship can be used to quickly estimate the initial impedance value during the simulation initialization phase, thus accelerating convergence.

[0026] In one embodiment, the measured airflow leakage characteristic value is determined based on the matching degree between the energy distribution of the measured electrical signal of the device under test and the preset structural transfer coefficient, including: Frequency domain analysis was performed on the measured electrical signals of the device under test to obtain the relative harmonic amplitude sequences of the inner electrode and the outer electrode. Based on the preset structural transmission coefficient, the relative harmonic amplitude sequence of the inner electrode and the relative harmonic amplitude sequence of the outer electrode are weighted and matched to obtain the structural vibration consistency coefficient. The measured characteristic value of airflow leakage is determined based on the structural vibration consistency coefficient.

[0027] Specifically, based on the structural vibration consistency coefficient, the measured airflow leakage characteristic values ​​are determined, including: The measured airflow leakage characteristic value is determined based on the difference between the preset value and the structural vibration consistency coefficient.

[0028] The preset value can be set according to the actual situation, for example, 1.

[0029] Receive input internal electrode time-domain voltage Internal electrode time-domain current and the time-domain voltage of the external electrode External electrode time-domain current The instantaneous power waveforms of the inner and outer electrodes are calculated separately, and frequency domain analysis is performed on these two power waveforms, i.e., Fast Fourier Transform (FFT) is executed to obtain the power spectral density in the frequency domain.

[0030] Extracting the DC component from the power spectral density ( ) and before The amplitude of the first-order AC harmonic component (e.g., taking...) ). Record the first internal electrode. The amplitude of the subharmonic is External electrode number The amplitude of the subharmonic is .

[0031] To ensure comparability of characteristics under different voltages, adaptive normalization is required. This is achieved by using the fundamental amplitude of each signal (…). This serves as the normalization benchmark. Specifically, to prevent calculation errors with a zero denominator at weak signals or anti-resonance points, a very small positive number is introduced into the denominator. (For example, take) The calculation formula for normalization is as follows: ; ; in, For the normalized internal electrode number The amplitude of the second harmonic. For the normalized external electrode, the first The amplitude of the subharmonic.

[0032] Through the above steps, the relative harmonic amplitude sequence of the internal electrode is constructed. relative harmonic amplitude sequence with external electrode ,in, , These two sequences eliminate the influence of absolute amplitude and purely reflect the relative energy distribution of each harmonic.

[0033] In an ideal structure with no airflow leakage, the harmonic responses of the inner and outer electrodes of the same order should follow a fixed structural coupling ratio; however, airflow leakage, as an incoherent loss, will disrupt this fixed coupling ratio.

[0034] Therefore, a set of harmonic nodes containing internal electrodes is constructed. and external electrode harmonic node set The bipartite graph structure is then calculated. Next, the internal electrode harmonic nodes are calculated. ( Characterizing the internal electrode (subharmonics) and external electrode harmonic nodes ( Characterizing the external electrode Matching weights between subharmonics .

[0035] Specifically, for harmonics of the same order (i.e. ): Weight This reflects the degree to which the current measured proportions approximate the inherent proportions of the structure. The formula for calculating the matching weight is: ; in, for Matching weights of harmonics of the same order, For the internal electrode The amplitude of the second harmonic. For the preset result transfer coefficient, For the external electrode The amplitude of the second harmonic. This is the structural tolerance constant (e.g., 0.1). The larger the value, the higher the algorithm's tolerance for bias. When the measured proportion... The closer to the inherent proportion At that time, matching weight The closer it is to 1 (full marks).

[0036] For trans-order harmonics (i.e.) Considering that structural nonlinearity is mainly manifested as energy transfer at the same frequency, in order to prevent the algorithm from incorrectly matching unrelated frequency components, a matching weight is forcibly set. (or a very small penalty value) This constraint ensures that the algorithm will not produce out-of-order matches that violate physical common sense.

[0037] Extract the first First harmonic (i.e.) The matching weights of the pairs are summed to obtain the total weight. Then the total weight Divide by the total number of harmonics (i.e.) The normalized structural vibration uniformity coefficients were obtained. , The range of values ​​is . The closer it is to 1, the more closely the energy distribution pattern of the current electrical signal matches the inherent pattern of the structure under leak-free conditions, meaning that the degree of airflow leakage is low. A significant decrease indicates a large amount of abnormal energy loss that cannot be explained by structural laws, meaning that the degree of airflow leakage is high.

[0038] Measured airflow leakage characteristic value The calculation formula is: ; It represents the degree to which it conforms to structural regularity, and its complement. Characterizing the degree to which the structure does not conform to the laws of motion, that is, the measured characteristic value of airflow leakage. It characterizes the degree of airflow leakage. It is a dimensionless scalar with a range of values ​​of . . The larger the value, the more the spectral distribution deviates from the inherent structure's shape, indicating a more severe airflow leakage. This eigenvalue, as the final output of the algorithm operator, will be returned to the caller for subsequent deviation analysis.

[0039] In one embodiment, the acoustic impedance boundary is calculated based on the measured slit opening value of the device under test, including: Obtain the measured slit opening value of the device under test under the target driving voltage; Obtain the dynamic viscosity constant of air, the width of the slit, the length of the slit, and the initial gap of the slit; The acoustic impedance boundary is determined based on the sum of the measured slit opening value and the initial slit gap, the product of the dynamic viscosity constant of air and the length of the slit, and the width of the slit.

[0040] To provide a reasonable starting point for subsequent closed-loop calibration and avoid iterative oscillations due to excessive initial value deviations, boundary initialization is required. Specifically, based on the driving conditions calculated in the simulation, a target voltage amplitude is designed. By examining the data mapping relationship established in the above steps, the target voltage amplitude can be obtained. Measured slit opening value Then, based on the parallel plate flow theory in fluid mechanics, the theoretical acoustic impedance at the slit is estimated. For a width of... , length is Height is The slit, its acoustic impedance With height It is inversely proportional to the cube of the equation. Calculate the initial acoustic impedance boundary. The calculation formula is: ; in, The aerodynamic viscosity constant is... The width of the slit. The length of the slit. This represents the initial gap of the slit, which is the initial height of the slit when no driving voltage is applied.

[0041] Through the above calculations, the geometric information from physical measurements is transformed into initial values ​​of boundary parameters that the simulation model can recognize. This not only gives the laser measurement data a clear purpose, but also ensures that the simulation starts from a physical point close to reality.

[0042] S12. Run the preset simulation model to obtain the virtual waveform under the acoustic impedance boundary, and extract the virtual airflow leakage characteristic value based on the virtual waveform.

[0043] The operator creates a finite element model of the piezoelectric MEMS loudspeaker in multiphysics simulation software (such as COMSOL Multiphysics or ANSYS), which is the preset simulation model. The model should include the geometric structure of the piezoelectric layer, silicon substrate layer, and air domain. The driving conditions for the simulation calculation are set, and a target voltage amplitude is selected. Target voltage amplitude It should belong to the driving voltage amplitude. Within the scanning range.

[0044] In one embodiment, a preset simulation model is run to obtain a virtual waveform under acoustic impedance boundaries, including: Run the preset simulation model and perform transient calculations under the current acoustic impedance boundary to obtain the virtual voltage and virtual current waveforms at the inner electrode port and the outer electrode port; wherein, the virtual waveforms include the virtual voltage and virtual current waveforms at the inner electrode port and the outer electrode port.

[0045] Specifically, virtual airflow leakage feature values ​​are extracted based on the virtual waveform, including: Frequency domain analysis is performed on the virtual waveform to generate virtual relative harmonic amplitude sequences for the inner electrode and the outer electrode, respectively. Based on the preset structural transmission coefficient, the virtual relative harmonic amplitude sequence of the inner electrode and the virtual relative harmonic amplitude sequence of the outer electrode are weighted and matched to obtain the virtual structural vibration consistency coefficient. The virtual airflow leakage characteristic value is determined based on the difference between the preset value and the vibration consistency coefficient of the virtual structure.

[0046] After setting the initial acoustic impedance boundary Then, the preset simulation model is started to perform transient calculations, outputting virtual voltage and current waveforms at the inner and outer electrode ports, respectively. These virtual waveforms reflect the electrical response behavior predicted by the model under the currently set acoustic impedance boundary.

[0047] To assess the deviation between the current model state and the actual state, the four sets of virtual waveform data were used as input parameters. Following the same steps as calculating the measured airflow leakage characteristic values, frequency domain analysis, relative harmonic amplitude sequence generation, weighted matching, and structural vibration consistency coefficient calculation were performed sequentially to obtain the virtual airflow leakage characteristic values. Specifically, frequency domain analysis was performed on the four sets of virtual waveform data to generate virtual relative harmonic amplitude sequences for the inner electrode and the outer electrode, respectively. Based on a preset structural transfer coefficient, weighted matching was performed on the virtual relative harmonic amplitude sequences of the inner and outer electrodes. Specifically, for harmonic pairs of the same order, the closeness between the virtual relative amplitude of the inner electrode and the product of the structural transfer coefficient and the virtual relative amplitude of the outer electrode was calculated, serving as the matching weight for that order harmonic pair. For harmonic pairs of different orders, the matching weight was set to zero. Then, in the bipartite graph structure, a matching scheme that maximizes the sum of the weights of all selected matching edges is found, and the maximum total weight is divided by the total number of harmonics to obtain the virtual structure vibration consistency coefficient. Based on the difference between 1 and the virtual structure vibration consistency coefficient, the virtual airflow leakage characteristic value is determined.

[0048] Perform the same frequency domain analysis and weighted matching calculation on the virtual waveform, remove the structural nonlinear components from the virtual signal, and finally return the virtual airflow leakage characteristic value of the current model. ,in, Represents the current iteration round, initially set to 0. The exact same algorithm processing flow is used for both the measured and virtual signals, ensuring... and They have completely consistent mathematical definitions and physical meanings, thus ensuring that the characteristic deviation between the two truly reflects the differences in physical boundary conditions, rather than algorithmic errors.

[0049] S13. Calculate the characteristic deviation between the measured airflow leakage characteristic value and the virtual airflow leakage characteristic value, and adjust the acoustic impedance boundary in the simulation model according to the direction and value of the characteristic deviation.

[0050] In one embodiment, calculating the characteristic deviation between the measured airflow leakage characteristic value and the virtual airflow leakage characteristic value includes: The difference between the virtual airflow leakage characteristic value and the measured airflow leakage characteristic value is calculated to obtain the characteristic deviation.

[0051] Specifically, the acoustic impedance boundary in the simulation model is adjusted according to the direction and value of the characteristic deviation, including: The direction of adjustment of the acoustic impedance boundary is determined by the sign of the characteristic deviation. The adjustment range is determined by the ratio of the absolute value of the characteristic deviation to the measured characteristic value of airflow leakage.

[0052] Specifically, the adjustment direction of the acoustic impedance boundary is determined based on the sign of the characteristic deviation, including: When the characteristic deviation is positive, increase the acoustic impedance boundary; When the characteristic deviation is negative, reduce the acoustic impedance boundary.

[0053] Calculate the characteristic deviation of the nth round. The calculation formula is: ;in, The virtual airflow leakage characteristic value obtained from the nth simulation round is... These are the measured characteristic values ​​of airflow leakage.

[0054] Next, based on the characteristic deviation The sign of the determines the direction of adjustment of the acoustic impedance boundary, specifically: like This indicates that the virtual airflow leakage characteristic value is too large, meaning the airflow leakage in the simulation model is more severe than in reality (the gap is too transparent). In this case, it is necessary to increase the acoustic impedance boundary. This increases flow resistance and suppresses leakage.

[0055] like This indicates that the virtual airflow leakage characteristic value is too small, meaning the simulation model is too enclosed. In this case, it is necessary to reduce the acoustic impedance boundary to simulate a more open leakage environment.

[0056] Given the acoustic impedance boundary The numerical range is extremely large (usually in) to (Between), simple linear adjustment is prone to numerical divergence or adjustment failure. Therefore, a logarithmic domain proportional adjustment strategy is adopted, updating at the order of magnitude level. The adjustment range of the acoustic impedance boundary is determined based on the ratio of the absolute value of the characteristic deviation to the measured airflow leakage characteristic value. The adjustment formula is: ; in, The acoustic impedance boundary for the (n+1)th simulation round. The acoustic impedance boundary for the nth simulation round, To adjust the step size coefficient and control the sensitivity of the adjustment. The larger the value, the greater the change in impedance in a single iteration, and the faster the convergence speed. However, an excessively large value may cause oscillations around the target value. The range of values ​​is to . This is the sign function. It is taken when the characteristic deviation is positive. This causes the logarithm of the impedance to increase, i.e., the acoustic impedance boundary to increase; when the characteristic deviation is negative, take The acoustic impedance boundary decreases. This is a limiting factor used to prevent runaway control during a single adjustment due to excessive initial deviation. For example, setting... This means that a single adjustment can change the value by a maximum of 0.5 orders of magnitude, ensuring the stability of the iteration process.

[0057] S14. Repeat steps S12 and S13 until the characteristic deviation meets the preset convergence condition, and output the sound pressure distribution data under the acoustic impedance boundary.

[0058] In one embodiment, the preset convergence condition includes: the ratio of the absolute value of the characteristic deviation to the measured airflow leakage characteristic value is less than a preset convergence threshold.

[0059] After updating the acoustic impedance boundary, it is necessary to determine whether the calibration process is complete. Check the ratio of the absolute value of the characteristic deviation to the measured airflow leakage characteristic value (i.e.,...) Is it less than the preset convergence threshold (e.g., the convergence threshold is 1)? ).

[0060] If the preset convergence condition is not met: let the iteration rounds... Utilizing the updated acoustic impedance boundary Return to step S12, rerun the simulation calculation and feature extraction, and enter the next loop.

[0061] If the preset convergence condition is met, it indicates that the virtual leakage behavior of the current simulation model has highly approximated the real device, meaning that the acoustic impedance boundary in the model has effectively represented the dynamic airflow turbulence effect in the real world. Stop the loop and lock the current acoustic impedance boundary. Read the sound pressure level (SPL) data of the entire sound field of the simulation model and the frequency response curve of the target observation point (e.g., at 10cm), and output the calibrated sound pressure distribution data as the test result to the user. Since this result is calculated under the boundary conditions aligned with leakage characteristics, it accurately reflects the real acoustic performance of the piezoelectric MEMS loudspeaker under high voltage and high leakage conditions.

[0062] This application utilizes a spectrum consistency decoupling algorithm to effectively address the technical challenge of conventional power testing failing to distinguish between structural losses and airflow losses. By eliminating harmonic energy that conforms to the inherent proportions of the structure, the extracted feature values ​​can specifically reflect the airtightness of the slit, avoiding overcompensation of simulation damping parameters due to misjudgment of piezoelectric material nonlinear losses, and significantly improving the physical relevance of feature extraction.

[0063] Furthermore, based on the aforementioned decoupling characteristics, the acoustic impedance boundary closed-loop calibration mechanism established in this application enables the simulation model to adaptively "learn" the real dynamic leakage behavior. Compared to traditional methods that rely on ideal fixed boundaries or expensive fluid-structure interaction calculations, this application successfully reproduces the low-frequency sound pressure attenuation phenomenon caused by the dynamic opening and closing of the slit by iteratively correcting the acoustic impedance boundary at extremely low computational cost. This significantly improves the agreement between the simulation prediction curve and the measured data across the entire frequency range (especially in the low-frequency large amplitude range), providing a reliable basis for optimizing the acoustic performance of the device.

[0064] This application also provides a sound pressure distribution simulation and testing system for a piezoelectric MEMS loudspeaker, such as Figure 2 As shown, the system includes: The determination module is used to determine the measured airflow leakage characteristic value based on the matching degree between the energy distribution of the measured electrical signal of the device under test and the preset structural transfer coefficient, and to calculate the acoustic impedance boundary based on the measured slit opening value of the device under test. The acquisition module is used to run a preset simulation model, acquire virtual waveforms under acoustic impedance boundaries, and extract virtual airflow leakage characteristic values ​​based on the virtual waveforms. The calculation module is used to calculate the characteristic deviation between the measured airflow leakage characteristic value and the virtual airflow leakage characteristic value, and adjust the acoustic impedance boundary in the simulation model according to the direction and value of the characteristic deviation. The iteration module is used to repeatedly call the acquisition module and the calculation module until the characteristic deviation meets the preset convergence condition and outputs the sound pressure distribution data under the acoustic impedance boundary.

[0065] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0066] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker, characterized in that, The method includes: S11. Based on the matching degree between the energy distribution of the measured electrical signal of the device under test and the preset structural transmission coefficient, determine the measured air leakage characteristic value, and calculate the acoustic impedance boundary based on the measured slit opening value of the device under test. S12. Run the preset simulation model to obtain the virtual waveform under the acoustic impedance boundary, and extract the virtual airflow leakage characteristic value based on the virtual waveform; wherein, the virtual waveform includes the virtual voltage waveform and virtual current waveform of the inner electrode port, and the virtual voltage waveform and virtual current waveform of the outer electrode port. S13. Calculate the characteristic deviation between the measured airflow leakage characteristic value and the virtual airflow leakage characteristic value, and adjust the acoustic impedance boundary in the simulation model according to the direction and value of the characteristic deviation. S14. Repeat steps S12 and S13 until the characteristic deviation meets the preset convergence condition, and output the sound pressure distribution data under the acoustic impedance boundary. The step of extracting virtual airflow leakage feature values ​​based on the virtual waveform includes: Frequency domain analysis is performed on the virtual waveform to generate virtual relative harmonic amplitude sequences for the inner electrode and the outer electrode, respectively. Based on the preset structural transmission coefficient, the virtual relative harmonic amplitude sequence of the inner electrode and the virtual relative harmonic amplitude sequence of the outer electrode are weighted and matched to obtain the virtual structural vibration consistency coefficient. The virtual airflow leakage characteristic value is determined based on the difference between the preset value and the vibration consistency coefficient of the virtual structure.

2. The method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker as described in claim 1, characterized in that, The step of determining the measured airflow leakage characteristic value based on the matching degree between the energy distribution of the measured electrical signal of the device under test and the preset structural transfer coefficient includes: Frequency domain analysis was performed on the measured electrical signals of the device under test to obtain the relative harmonic amplitude sequences of the inner electrode and the outer electrode. Based on the preset structural transmission coefficient, the relative harmonic amplitude sequence of the inner electrode and the relative harmonic amplitude sequence of the outer electrode are weighted and matched to obtain the structural vibration consistency coefficient. The measured characteristic value of airflow leakage is determined based on the structural vibration consistency coefficient.

3. The method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker as described in claim 2, characterized in that, The determination of the measured airflow leakage characteristic value based on the structural vibration consistency coefficient includes: The measured airflow leakage characteristic value is determined based on the difference between the preset value and the structural vibration consistency coefficient.

4. The method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker as described in claim 1, characterized in that, The calculation of the acoustic impedance boundary based on the measured slit opening value of the device under test includes: Obtain the measured slit opening value of the device under test under the target driving voltage; Obtain the dynamic viscosity constant of air, the width of the slit, the length of the slit, and the initial gap of the slit; The acoustic impedance boundary is determined based on the sum of the measured slit opening value and the initial slit gap, the product of the dynamic viscosity constant of air and the length of the slit, and the width of the slit.

5. The method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker as described in claim 1, characterized in that, The process of running a preset simulation model to obtain virtual waveforms under acoustic impedance boundaries includes: Run the preset simulation model and perform transient calculations under the current acoustic impedance boundary to obtain the virtual voltage and virtual current waveforms at the inner electrode port and the outer electrode port.

6. The method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker as described in claim 1, characterized in that, The characteristic deviation between the calculated measured airflow leakage characteristic value and the virtual airflow leakage characteristic value includes: The difference between the virtual airflow leakage characteristic value and the measured airflow leakage characteristic value is calculated to obtain the characteristic deviation.

7. The method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker as described in claim 6, characterized in that, The adjustment of the acoustic impedance boundary in the simulation model based on the direction and value of the characteristic deviation includes: The direction of adjustment of the acoustic impedance boundary is determined by the sign of the characteristic deviation. The adjustment range is determined by the ratio of the absolute value of the characteristic deviation to the measured characteristic value of airflow leakage.

8. The method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker as described in claim 7, characterized in that, The method of determining the adjustment direction of the acoustic impedance boundary based on the sign of the characteristic deviation includes: When the characteristic deviation is positive, increase the acoustic impedance boundary; When the characteristic deviation is negative, reduce the acoustic impedance boundary.

9. The method for simulating and testing the sound pressure distribution of a piezoelectric MEMS loudspeaker as described in claim 1, characterized in that, The preset convergence conditions include: the ratio of the absolute value of the characteristic deviation to the measured airflow leakage characteristic value is less than the preset convergence threshold.

Citation Information

Patent Citations

  • Method of Controlling Sound Reproduction of Enclosure Mounted Loudspeakers

    US20150030169A1

  • Load change diagnostics for acoustic devices and methods

    US20190327563A1