A thermal coupling decoupling measurement method based on a composite sensitive structure

Abstract

The application discloses a kind of thermal coupling decoupling measurement methods based on composite sensitive structure, it is related to extreme environment sensing measurement technical field, provide composite sensitive structure: composite sensitive structure from bottom to top includes substrate, thermoelectric pile film formed on substrate, heat insulation layer formed on thermoelectric pile film, and piezoelectric film formed on heat insulation layer;Dynamic response characteristic matrix of composite sensitive structure is constructed: the thermal coupling coefficient between piezoelectric film and thermoelectric pile film when pressure and thermal excitation act simultaneously is calculated, and dynamic response characteristic matrix is constituted by amplitude-frequency response coefficient, phase-frequency response coefficient and thermal coupling coefficient.The application adopts the technical route of composite sensitive structure, dynamic characteristic matrix and frequency domain decoupling, solves the problem that traditional method is difficult to separate thermal coupling signal in real time from measurement principle, realizes the synchronous measurement of dynamic pressure and transient temperature under extreme environment.

Description

Technical Field

[0001] This invention relates to the field of extreme environment sensing and measurement technology, specifically to a thermo-coupling and decoupling measurement method based on a composite sensitive structure. Background Technology

[0002] In extreme environments such as internal combustion engine knock, hypersonic airflow, and aero-engine combustion chambers, pressure and temperature parameters are not only extremely high (pressures can reach tens of megapascals and temperatures can reach thousands of Kelvins), but also change at extremely rapid rates (on the order of microseconds or even nanoseconds). Furthermore, a strong thermo-mechanical coupling effect exists between the two. Accurate and synchronous measurement of dynamic pressure waveforms and transient temperature changes under these extreme conditions is of crucial scientific and engineering value for revealing complex flow mechanisms, verifying numerical simulation results, optimizing thermal protection design, and realizing active engine control.

[0003] In existing technologies, sensing and measurement in high-temperature and high-pressure environments mainly face the following technical challenges: traditional single-parameter sensors struggle to simultaneously acquire pressure and temperature information; existing composite sensors suffer from signal interference and decoupling difficulties; while fiber optic Fabry-Perot composite sensors can achieve dual-parameter measurement of temperature and pressure, their structure typically employs a fusion splicing of hollow and solid optical fibers, achieving decoupling through controlling the cavity length difference. This approach is limited by the principle of optical interference, resulting in limited response to extremely rapid pressure fluctuations (such as shock waves and detonation waves), and the refractive index change of the fiber material at high temperatures introduces additional measurement errors.

[0004] In summary, existing technologies lack an effective method for simultaneously measuring dynamic pressure and transient temperature under extreme conditions and for separating thermo-mechanical coupling signals at their source.

[0005] Patent CN111929145B discloses a method and device for measuring the composite field of high-temperature thermo-mechanical coupling properties of metals based on the virtual field method. The patent combines two non-contact full-field measurement technologies, digital image correlation and infrared thermal imaging, and overcomes the assumptions of uniform temperature distribution and uniform deformation distribution that are the basis of conventional thermo-mechanical coupling property testing methods.

[0006] The aforementioned patents reduce the constraints of experimental testing of thermo-coupling properties, but cannot simultaneously measure dynamic pressure and transient temperature under extreme conditions, nor can they fundamentally separate thermo-coupling signals.

[0007] To this end, this application proposes a thermo-coupling decoupling measurement method based on a composite sensitive structure that synchronously reconstructs the real dynamic pressure waveform and transient temperature change at the same spatiotemporal point. Summary of the Invention

[0008] The purpose of this invention is to provide a thermo-mechanical coupling decoupling measurement method based on a composite sensitive structure to solve the technical problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a thermo-mechanical coupling decoupling measurement method based on a composite sensitive structure, the method comprising the following steps:

[0010] A composite sensing structure is provided: the composite sensing structure includes, from bottom to top, a substrate, a thermopile thin film formed on the substrate, a heat insulation layer formed on the thermopile thin film, and a piezoelectric thin film formed on the heat insulation layer;

[0011] Constructing the dynamic response feature matrix of the composite sensitive structure: At the same time, a step pressure excitation of known amplitude and a pulsed laser thermal excitation of known density are simultaneously applied to the surface of the composite sensitive structure. The first dynamic response signal of the piezoelectric film in response to the step pressure excitation and the pulsed laser thermal excitation, and the second dynamic response signal of the thermopile film in response to the step pressure excitation and the pulsed laser thermal excitation are simultaneously acquired. Fourier transforms are performed on the first dynamic response signal and the second dynamic response signal respectively to obtain the amplitude frequency response coefficient and phase frequency response coefficient of the piezoelectric film and the thermopile film at different frequencies in response to pressure excitation, and the amplitude frequency response coefficient and phase frequency response coefficient of the thermopile film in response to thermal excitation. The thermo-coupling coefficient between the piezoelectric film and the thermopile film when pressure and thermal excitation are applied simultaneously is calculated. The dynamic response feature matrix is ​​composed of the amplitude frequency response coefficient, the phase frequency response coefficient, and the thermo-coupling coefficient.

[0012] Preferably, in the composite sensitive structure, the piezoelectric thin film is prepared using aluminum nitride or zinc oxide material and deposited on the upper surface of the heat insulation layer by magnetron sputtering process. A first interdigitated electrode is also prepared on the upper surface of the piezoelectric thin film.

[0013] The thermopile film is composed of multiple p-type polycrystalline silicon thermocouple arms and n-type polycrystalline silicon thermocouple arms connected in series. The hot ends of the p-type polycrystalline silicon thermocouple arms and n-type polycrystalline silicon thermocouple arms are located directly below the heat insulation layer, and the cold ends are located in the area of ​​the substrate that is not covered by the heat insulation layer. The cold ends of the thermopile film are led out through the second interdigitated electrode.

[0014] The heat insulation layer is a silica aerogel film or a porous alumina film, with a thickness of 3-5 times that of the piezoelectric film.

[0015] Preferably, when constructing the dynamic response feature matrix, the step pressure excitation is generated by a shock tube, the outlet of the shock tube is sealed to the surface of the composite sensitive structure, the rise time of the step pressure is less than 1 microsecond, and the pressure amplitude is calibrated by changing the gas pressure difference between the driving section and the driven section of the shock tube.

[0016] The pulsed laser thermal excitation is generated by a semiconductor laser. The laser wavelength is selected to be in the band where the piezoelectric thin film absorptivity is greater than 70%. The laser spot completely covers the surface of the composite sensitive structure. The rise time of the laser pulse is less than 10 nanoseconds. The laser power density is calibrated in real time by an optical power meter. The time difference between the laser pulse and the step pressure excitation reaching the surface of the composite sensitive structure is less than 10 nanoseconds.

[0017] Preferably, during actual measurement, the first mixed signal output by the piezoelectric thin film in the composite sensitive structure and the second mixed signal output by the thermopile thin film are simultaneously acquired. The first mixed signal and the second mixed signal are respectively subjected to Fourier transform to the frequency domain to obtain the first frequency domain mixed signal and the second frequency domain mixed signal. A frequency domain decoupling matrix is ​​constructed based on the dynamic response feature matrix. The frequency domain decoupling matrix is ​​used to perform blind source separation on the first frequency domain mixed signal and the second frequency domain mixed signal to obtain the frequency domain temperature signal of the frequency domain pressure signal. The frequency domain pressure signal and the frequency domain temperature signal are respectively subjected to inverse Fourier transform to restore the dynamic pressure time domain waveform and the transient temperature time domain change curve that are completely synchronized with the first mixed signal and the second mixed signal in time.

[0018] Preferably, the method for calculating the thermodynamic coupling coefficient is as follows:

[0019] First, under the condition of only step pressure excitation, the first pressure response signal of the piezoelectric film and the second pressure response signal of the thermopile film are obtained respectively.

[0020] Secondly, under the condition of only pulsed laser thermal excitation, the first thermal response signal of the piezoelectric thin film and the second thermal response signal of the thermopile thin film were obtained respectively.

[0021] Then, step pressure excitation and pulsed laser thermal excitation are applied simultaneously to obtain the first total response signal and the first thermal response signal of the piezoelectric film, respectively. The difference is used as the thermo-mechanical coupling response quantity of the piezoelectric film. The second total response signal is subtracted from the second pressure response signal and the second thermal response signal, and the difference is used as the thermo-mechanical coupling response quantity of the thermopile film. The thermo-mechanical coupling response quantities of the piezoelectric film and the thermopile film are normalized in the frequency domain to obtain the thermo-mechanical coupling coefficient matrix.

[0022] Preferably, the construction process of the frequency domain decoupling matrix is ​​as follows: the amplitude frequency response coefficient and phase frequency response coefficient in the dynamic response characteristic matrix are combined into a complex form pressure transfer function and heat transfer function, and the thermodynamic coupling coefficient is superimposed on the pressure transfer function and heat transfer function in complex form to form a 2×2 complex transfer function matrix; the complex transfer function matrix is ​​inverted at each discrete frequency point to obtain the frequency domain decoupling matrix.

[0023] Preferably, the blind source separation process is as follows: the first frequency domain mixed signal and the second frequency domain mixed signal are combined to form a 2×1 frequency domain mixed vector, and the frequency domain decoupling matrix is ​​multiplied by the frequency domain mixed vector to obtain a 2×1 frequency domain source vector. The two elements in the frequency domain source vector are the frequency domain pressure signal and the frequency domain temperature signal.

[0024] Preferably, during the construction of the dynamic response feature matrix, multiple calibration experiments are repeated for different combinations of pressure and thermal excitation amplitudes to obtain multiple sets of dynamic response feature matrices under different amplitude ranges. In actual measurement, the dynamic response feature matrix of the corresponding amplitude range is adaptively selected to construct the frequency domain decoupling matrix based on the real-time amplitude of the first mixed signal and the second mixed signal, so as to eliminate the nonlinear response error introduced by the composite sensitive structure when measuring in a large dynamic range.

[0025] Preferably, a heat-reflective layer is further provided between the upper surface of the heat insulation layer and the lower surface of the piezoelectric film. The heat-reflective layer is a gold or silver film with a thickness of 100 nanometers to 500 nanometers. It is used to reflect the infrared portion that passes through the piezoelectric film during pulsed laser thermal excitation back to the piezoelectric film, thereby improving the absorption efficiency of the piezoelectric film for laser thermal excitation and reducing the direct radiation of heat to the thermopile film.

[0026] Preferably, after restoring the dynamic pressure time-domain waveform and transient temperature time-domain change curve, a time-domain calibration step is also included: the restored dynamic pressure time-domain waveform and transient temperature time-domain change curve are used as inputs and substituted into the pre-established composite sensitive structure thermo-coupling simulation model to calculate the simulated piezoelectric film output signal and thermopile film output signal in a forward manner. The simulated output signal is then compared with the actual measured first mixed signal and second mixed signal. If the root mean square error between the two exceeds a preset threshold, an iterative deconvolution algorithm is used to correct the frequency domain decoupling matrix, and blind source separation and inverse Fourier transform are performed again until the root mean square error is less than the preset threshold.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] 1. This invention adopts a composite sensitive structure, dynamic feature matrix and frequency domain decoupling technical approach, which solves the problem of traditional methods being unable to separate thermo-coupled signals in real time from the measurement principle, and realizes the synchronous measurement of dynamic pressure and transient temperature under extreme environments;

[0029] 2. The composite sensitive structure designed in this invention uses the longitudinal stacking of piezoelectric thin film and thermopile thin film, combined with the thermal isolation effect of the intermediate heat insulation layer, to enable the two sensitive materials to have dominant sensitivity to pressure and thermal excitation respectively, while reducing mutual interference through structural design;

[0030] 3. This invention obtains the dynamic response characteristic matrix of the composite sensitive structure by simultaneously applying step pressure excitation and pulsed laser thermal excitation through shock tube and pulsed laser. This matrix comprehensively describes the amplitude-frequency characteristics, phase-frequency characteristics and coupling characteristics of the two sensitive materials, providing an accurate mathematical model basis for frequency domain decoupling.

[0031] 4. This invention employs a frequency-domain blind source separation algorithm, which utilizes the difference in response characteristics of two sensitive materials to different excitations to achieve real-time decoupling of mixed signals. The resulting dynamic pressure waveform and transient temperature change curve have complete time-domain synchronization. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the dynamic response feature matrix construction process of the present invention;

[0033] Figure 2 This is a schematic diagram of the actual measurement time-frequency domain decoupling and signal restoration process of the present invention;

[0034] Figure 3 This is a schematic diagram illustrating the principle of adaptive selection of dynamic response feature matrices for different amplitude ranges in this invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] Example 1

[0037] Please see Figure 1 , Figure 2 and Figure 3 A thermo-coupling and decoupling measurement method based on a composite sensitive structure, and the fabrication of the composite sensitive structure:

[0038] The substrate is a double-sided polished high-resistivity monocrystalline silicon wafer with a thickness of 500 micrometers and a resistivity greater than 1000 Ω·cm to reduce interference from the substrate to the sensor signal. The thermopile thin film is fabricated on the upper surface of the substrate. The specific fabrication process is as follows: First, a silicon dioxide insulating layer with a thickness of 0.5 micrometers is grown on the silicon substrate by thermal oxidation; then, a polycrystalline silicon thin film with a thickness of 0.8 micrometers is deposited by low-pressure chemical vapor deposition; then, p-type polycrystalline silicon thermocouple arms and n-type polycrystalline silicon thermocouple arms are formed by ion implantation and annealing processes, respectively, with a doping concentration of 1×10⁻⁶. 20 cm -3The polycrystalline silicon thin film is patterned using photolithography and etching processes to form a thermopile structure consisting of multiple p-type and n-type polycrystalline silicon thermocouple arms connected in series. The width of each thermocouple arm is 10 micrometers and the length is 200 micrometers. Finally, metal electrodes are deposited at both ends of the thermocouple arms. The hot end electrode is located directly below the subsequent thermal insulation layer, and the cold end electrode is located in the area of ​​the substrate not covered by the thermal insulation layer. The cold end electrode is led out through a second interdigitated electrode.

[0039] The heat insulation layer is fabricated on top of the thermopile film, covering the hot end region of the thermopile. The heat insulation layer is made of silica aerogel film, which is prepared by sol-gel process combined with supercritical drying technology. The thickness is 3 micrometers. The porosity of silica aerogel is greater than 90% and the thermal conductivity is less than 0.05 W / m·K. The thickness of the heat insulation layer is designed to be 4 times the thickness of the piezoelectric film to achieve the best thermal insulation effect.

[0040] The piezoelectric thin film is fabricated on the upper surface of the heat insulation layer. The piezoelectric thin film is made of aluminum nitride material and is deposited by reactive magnetron sputtering process with a sputtering power of 300W, a nitrogen-argon flow ratio of 1:1, a deposition temperature of 300℃, and a film thickness of 0.75 micrometers. The aluminum nitride thin film has a high piezoelectric coefficient (about 5pC / N) and good high temperature stability (can work above 600℃). On the upper surface of the piezoelectric thin film, the first interdigital electrode is fabricated by photolithography and lift-off process. The electrode material is a platinum-titanium composite layer with a thickness of 200nm / 20nm. The finger width and finger spacing of the interdigital electrode are both 10 micrometers.

[0041] Example 2

[0042] Please see Figure 1 , Figure 2 and Figure 3 A thermo-coupling and decoupling measurement method based on a composite sensitive structure, the process of constructing the dynamic response feature matrix includes the following sub-steps:

[0043] A calibration experimental system was constructed, comprising a shock tube, a pulsed semiconductor laser, a high-speed data acquisition card, and a synchronization controller. Both the driving and driven sections of the shock tube were 2 meters long, and the shock tube outlet was sealed to the surface of the composite sensitive structure. An 808nm wavelength laser was selected as the pulsed semiconductor laser, at which the absorptivity of the aluminum nitride piezoelectric film is approximately 75%. The high-speed data acquisition card was a 4-channel synchronous acquisition card with a sampling rate of 2.5GS / s, a modulus resolution of 12 bits, and a channel synchronization deviation of less than 30 picoseconds. The synchronization controller was used to precisely control the synchronization between the shock tube film breaking moment and the laser pulse triggering moment.

[0044] Pure pressure excitation calibration is performed under the condition of only step pressure excitation. The pressure difference between the driving section and the driven section of the shock tube is adjusted to generate a step pressure with a rise time of less than 0.8 microseconds and an amplitude of 1 MPa. When the step pressure reaches the surface of the composite sensitive structure, the first pressure response signal output by the piezoelectric film and the second pressure response signal output by the thermopile film are collected simultaneously. The experiment is repeated 10 times, and the collected signals are averaged to reduce random noise.

[0045] Pure thermal excitation calibration was performed under conditions of only pulsed laser thermal excitation. The laser output power was adjusted to achieve a laser power density of 10 kW / cm². 2 The laser pulse width is 50 nanoseconds and the rise time is less than 8 nanoseconds. When the laser pulse reaches the surface of the composite sensitive structure, the first thermal response signal output by the piezoelectric film and the second thermal response signal output by the thermopile film are collected simultaneously. The experiment is repeated 10 times and the signals are averaged.

[0046] Simultaneous thermal excitation calibration was performed by adjusting the synchronous controller so that the step pressure excitation and pulsed laser thermal excitation reached the surface of the composite sensitive structure almost simultaneously, with the time difference between the two arrivals controlled within 5 nanoseconds. The first total response signal output by the piezoelectric thin film and the second total response signal output by the thermopile thin film were collected synchronously. The experiment was repeated 10 times, and the signals were averaged.

[0047] Signal preprocessing and Fourier transform: DC component removal is performed on all acquired signals, and Hanning windows are applied to reduce spectral leakage. Fast Fourier transform is then performed on the windowed signals to obtain the spectrum of each signal.

[0048] The amplitude-frequency response coefficient and phase-frequency response coefficient are calculated as follows: For a piezoelectric film, the amplitude-frequency response coefficient under pure pressure excitation is the spectral amplitude of the first pressure response signal divided by the spectral amplitude of the known step pressure, and the phase-frequency response coefficient is the spectral phase of the first pressure response signal minus the spectral phase of the step pressure. Similarly, the amplitude-frequency and phase-frequency response coefficients of the piezoelectric film under pure thermal excitation can be calculated. The same method is used to calculate the amplitude-frequency and phase-frequency response coefficients of the thermopile film under pure pressure and pure thermal excitation.

[0049] To calculate the thermodynamic coupling coefficients, firstly, the spectra of the first pressure response signal and the first thermal response signal are subtracted from the spectrum of the first total response signal to obtain the thermodynamic coupling response spectrum of the piezoelectric film; the spectra of the second pressure response signal and the second thermal response signal are subtracted from the spectrum of the second total response signal to obtain the thermodynamic coupling response spectrum of the thermopile film. Then, the above coupling response spectra are normalized in the frequency domain: the coupling response spectrum of the piezoelectric film is divided by the amplitude of the spectrum of the first pressure response signal to obtain the coupling coefficient of the piezoelectric film; the coupling response spectrum of the thermopile film is divided by the amplitude of the spectrum of the second pressure response signal to obtain the coupling coefficient of the thermopile film. These coupling coefficients constitute a 2×2 thermodynamic coupling coefficient matrix.

[0050] A dynamic response characteristic matrix is ​​constructed by combining the calculated amplitude-frequency response coefficients, phase-frequency response coefficients, and thermo-coupling coefficients to form a complete dynamic response characteristic matrix. For each discrete frequency point, this matrix contains four complex elements: H_pp(f) (the transfer function of the piezoelectric film to pressure), H_pt(f) (the transfer function of the piezoelectric film to heat), H_tp(f) (the transfer function of the thermopile film to pressure), H_tt(f) (the transfer function of the thermopile film to heat), and a coupling correction term C(f). In practice, the coupling correction term can be incorporated into the above transfer functions to form a corrected transfer function matrix.

[0051] Example 3

[0052] Please see Figure 1 , Figure 2 and Figure 3 A thermo-coupling decoupling measurement method based on a composite sensing structure, wherein the actual measurement and signal decoupling process includes the following sub-steps:

[0053] The composite sensing structure is installed in the extreme environment under test (such as the combustion chamber wall of an internal combustion engine). During the actual measurement, the first mixed signal x1(t) output by the piezoelectric thin film and the second mixed signal x2(t) output by the thermopile thin film are acquired simultaneously. The acquisition parameters are consistent with those used during calibration: sampling rate 2.5 GS / s, 12-bit resolution.

[0054] The acquired mixed signals x1(t) and x2(t) are preprocessed by removing DC and windowing, and then FFT is performed to transform them to the frequency domain to obtain the first frequency domain mixed signal X1(f) and the second frequency domain mixed signal X2(f).

[0055] Based on the dynamic response feature matrix constructed in step two, a frequency domain decoupling matrix is ​​constructed. Specifically, the transfer function at each frequency point f is combined into a 2×2 complex transfer function matrix H(f):

[0056] H(f)=[H_pp(f)+C_pp(f),H_pt(f)+C_pt(f);

[0057] H_tp(f)+C_tp(f),H_tt(f)+C_tt(f)];

[0058] Where C_pp(f), C_pt(f), C_tp(f), and C_tt(f) are correction terms derived from the thermal coupling coefficient. The frequency domain decoupling matrix H_inv(f) is obtained by inverting H(f) at each frequency point f.

[0059] Blind source separation is performed. X1(f) and X2(f) are combined into a frequency domain mixing vector X(f)=[X1(f); X2(f)]. The frequency domain decoupling matrix H_inv(f) is multiplied by X(f) on the left to obtain the frequency domain source vector Y(f)=[Y_p(f); Y_t(f)]=H_inv(f)·X(f), where Y_p(f) is the frequency domain pressure signal and Y_t(f) is the frequency domain temperature signal.

[0060] The frequency domain pressure signal Y_p(f) and the frequency domain temperature signal Y_t(f) are filtered. A frequency sampling filter based on the minimum phase method is used. The passband frequency range is determined to be 10kHz to 100MHz based on the frequency range with a signal-to-noise ratio higher than 30dB in the dynamic response characteristic matrix, and the stopband attenuation is designed to be 65dB. The filtered frequency domain signals are Y_p_filtered(f) and Y_t_filtered(f).

[0061] The filtered frequency domain signal is subjected to inverse Fourier transform (IFFT) to restore the dynamic pressure time domain waveform p(t) and transient temperature time domain change curve T(t), which are completely synchronized with the original mixed signals x1(t) and x2(t) in time.

[0062] This completes the entire process of separating the real dynamic pressure and transient temperature from the mixed sensor signals.

[0063] Example 4

[0064] Please see Figure 1 , Figure 2 and Figure 3 A thermo-coupling and decoupling measurement method based on a composite sensitive structure is proposed in this embodiment. This embodiment optimizes the method for large dynamic range measurements based on Embodiment 1. The main improvements lie in the segmented calibration and adaptive selection of the dynamic response feature matrix.

[0065] When constructing the dynamic response characteristic matrix, multiple calibration experiments were repeatedly conducted for different combinations of pressure and thermal excitation amplitudes. Specifically, five pressure amplitude ranges were set: 0.1-0.5 MPa, 0.5-1 MPa, 1-2 MPa, 2-5 MPa, and 5-10 MPa; and five thermal excitation power density ranges: 1-5 kW / cm². 2 5-10kW / cm 2 10-20kW / cm 2 20-50kW / cm 2 50-100kW / cm 2 By combining the pressure amplitude and thermal excitation amplitude in pairs, a total of 25 dynamic response characteristic matrices under calibration conditions were obtained, denoted as M. ij , where i represents the pressure amplitude range number and j represents the thermal excitation amplitude range number.

[0066] In actual measurement, a real-time amplitude monitoring step is added: while acquiring the first mixed signal x1(t) and the second mixed signal x2(t), the real-time root mean square amplitude of the signal is calculated. Based on the preset amplitude interval division threshold, the pressure amplitude interval i0 and the thermal excitation amplitude interval j0 of the current signal are automatically determined. Then, a corresponding set M_i0j0 is selected from the 25 stored sets of dynamic response feature matrices to construct the subsequent frequency domain decoupling matrix.

[0067] The other steps in this embodiment are the same as in Embodiment 1. By segmented calibration and adaptive selection, this embodiment can effectively compensate for the nonlinear response error that may be introduced by the composite sensitive structure during large dynamic range measurement, thereby improving measurement accuracy.

[0068] Example 5

[0069] Please see Figure 1 , Figure 2 and Figure 3 A thermo-coupling and decoupling measurement method based on a composite sensitive structure is proposed, which optimizes the design of the composite sensitive structure based on Example 1.

[0070] In this embodiment, a heat-reflective layer is added between the upper surface of the heat insulation layer and the lower surface of the piezoelectric film. The heat-reflective layer is a gold film, prepared by electron beam evaporation, with a thickness of 300 nanometers. The gold film has a reflectivity of over 95% in the 808 nm wavelength band, which can reflect the pulsed laser energy passing through the aluminum nitride piezoelectric film back to the piezoelectric film, increasing the absorption efficiency of the piezoelectric film for laser thermal excitation from 75% to approximately 94%. At the same time, this gold film layer also acts as a thermal barrier, further reducing the direct radiation conduction of heat to the thermopile film and reducing the thermal interference experienced by the thermopile film.

[0071] The fabrication process was adjusted as follows: After the thermal insulation layer was fabricated, a gold film pattern was first deposited on the upper surface of the thermal insulation layer using photolithography and electron beam evaporation, followed by the deposition of a piezoelectric thin film. Because the gold film has good conductivity, care must be taken to avoid short circuits between the electrode and the gold film when fabricating the first interdigital electrode on the piezoelectric thin film.

[0072] The other steps in this embodiment are the same as in Embodiment 1. By adding a heat-reflective layer, this embodiment improves the thermal excitation efficiency and enhances the thermal isolation effect, which helps to further improve the decoupling accuracy.

[0073] Example 6

[0074] Please see Figure 1 , Figure 2 and Figure 3 A thermo-coupling decoupling measurement method based on a composite sensitive structure is proposed. This embodiment adds a time-domain calibration step as a post-processing optimization based on Embodiment 1.

[0075] After reconstructing the dynamic pressure time-domain waveform p(t) and the transient temperature time-domain change curve T(t), the following calibration steps are performed:

[0076] Sub-step 1. Establish a thermo-mechanical coupling simulation model of the composite sensitive structure. This model is based on the finite element method and takes the geometric dimensions and material parameters (density, specific heat capacity, thermal conductivity, piezoelectric coefficient, Seebeck coefficient, etc.) of the composite sensitive structure as input. It can forward calculate the output signals of the piezoelectric film and the thermopile film under given pressure p(t) and temperature T(t) excitation.

[0077] Sub-step 2. Take the restored p(t) and T(t) as inputs and substitute them into the simulation model to calculate the simulated piezoelectric film output signal x1_sim(t) and thermopile film output signal x2_sim(t) in the forward calculation.

[0078] Sub-step 3. Compare the simulated output signals x1_sim(t) and x2_sim(t) with the actual measured first mixed signal x1(t) and second mixed signal x2(t), and calculate the root mean square error (RMSE). If the RMSE is greater than a preset threshold (e.g., 2% of full scale), proceed to sub-step 4 for correction; if the RMSE is less than or equal to the threshold, the decoupling result is considered to meet the accuracy requirements, and p(t) and T(t) are output as the final result.

[0079] Sub-step 4. The frequency domain decoupling matrix H_inv(f) is corrected using an iterative deconvolution algorithm. The correction method is as follows: Calculate the error spectrum E(f) = X(f) - X_sim(f) between the actual signal and the simulated signal, multiply the error spectrum by a relaxation factor α (α ranges from 0.1 to 0.3), and then superimpose it onto the original frequency domain source vector, i.e., Y_new(f) = Y(f) + α·H_inv(f)·E(f). The new frequency domain decoupling matrix H_inv_new(f) is then obtained from Y_new(f), or the new p_new(t) and T_new(t) can be directly obtained from Y_new(f) via IFFT.

[0080] Substitute the corrected p_new(t) and T_new(t) back into the simulation model and repeat sub-steps 2 to 4 until the RMSE is less than the preset threshold or the maximum number of iterations is reached (e.g., 10 times).

[0081] The other steps in this embodiment are the same as in Embodiment 1. By introducing time-domain calibration and iterative optimization, this embodiment can further improve measurement accuracy, making it particularly suitable for scientific research scenarios with extremely high accuracy requirements.

[0082] Comparative Example 1: A traditional discrete sensor scheme was used, consisting of a piezoelectric pressure sensor (response time approximately 2 microseconds) and a thin-film thermocouple temperature sensor (response time approximately 50 microseconds) installed near the measurement point. Measurement results were compared under the same shock tube dynamic testing conditions. The results show that the pressure sensor output in Comparative Example 1 exhibits significant thermally induced drift (error approximately 15%), the temperature sensor fails to capture the rapid temperature rise at the moment of shock wave arrival (the rising edge is smoothed), and due to the different installation positions of the two sensors, the measured pressure and temperature waveforms cannot be precisely correlated in time.

[0083] Comparative Example 2: A static calibration and polynomial fitting decoupling method was adopted. Static calibration of the composite sensitive structure was performed in a static pressure tank and temperature control chamber to establish a polynomial mapping relationship between pressure, temperature, and output signals. This mapping relationship was then used to decouple the dynamic mixed signals. Results showed that this method performed well under static conditions, but in dynamic testing (shock tube rise time 1 microsecond), the decoupled pressure waveform exhibited significant distortion, and the temperature waveform showed spurious oscillations. This is because static calibration cannot reflect the frequency domain response characteristics of the sensitive material under dynamic excitation, leading to the failure of high-frequency component decoupling.

[0084] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A thermo-mechanical coupling decoupling measurement method based on a composite sensitive structure, characterized in that: The method includes the following steps: A composite sensing structure is provided: the composite sensing structure includes, from bottom to top, a substrate, a thermopile thin film formed on the substrate, a heat insulation layer formed on the thermopile thin film, and a piezoelectric thin film formed on the heat insulation layer; Constructing the dynamic response feature matrix of the composite sensitive structure: At the same time, a step pressure excitation of known amplitude and a pulsed laser thermal excitation of known density are simultaneously applied to the surface of the composite sensitive structure. The first dynamic response signal of the piezoelectric film in response to the step pressure excitation and the pulsed laser thermal excitation, and the second dynamic response signal of the thermopile film in response to the step pressure excitation and the pulsed laser thermal excitation are simultaneously acquired. Fourier transforms are performed on the first dynamic response signal and the second dynamic response signal respectively to obtain the amplitude frequency response coefficient and phase frequency response coefficient of the piezoelectric film and the thermopile film at different frequencies in response to pressure excitation, and the amplitude frequency response coefficient and phase frequency response coefficient of the thermopile film in response to thermal excitation. The thermo-coupling coefficient between the piezoelectric film and the thermopile film when pressure and thermal excitation are applied simultaneously is calculated. The dynamic response feature matrix is ​​composed of the amplitude frequency response coefficient, the phase frequency response coefficient, and the thermo-coupling coefficient.

2. The thermo-coupling and decoupling measurement method based on a composite sensitive structure according to claim 1, characterized in that: In the composite sensitive structure, the piezoelectric film is made of aluminum nitride or zinc oxide material and is deposited on the upper surface of the heat insulation layer by magnetron sputtering. The upper surface of the piezoelectric film is also prepared with a first interdigitated electrode. The thermopile film is composed of multiple p-type polycrystalline silicon thermocouple arms and n-type polycrystalline silicon thermocouple arms connected in series. The hot ends of the p-type polycrystalline silicon thermocouple arms and n-type polycrystalline silicon thermocouple arms are located directly below the heat insulation layer, and the cold ends are located in the area of ​​the substrate that is not covered by the heat insulation layer. The cold ends of the thermopile film are led out through the second interdigitated electrode. The heat insulation layer is a silica aerogel film or a porous alumina film, with a thickness of 3-5 times that of the piezoelectric film.

3. The thermo-mechanical coupling decoupling measurement method based on a composite sensitive structure according to claim 1, characterized in that: When constructing the dynamic response feature matrix, the step pressure excitation is generated by the shock tube, the outlet of the shock tube is sealed to the surface of the composite sensitive structure, the rise time of the step pressure is less than 1 microsecond, and the pressure amplitude is calibrated by changing the gas pressure difference between the driving section and the driven section of the shock tube. The pulsed laser thermal excitation is generated by a semiconductor laser. The laser wavelength is selected to be in the band where the piezoelectric thin film absorptivity is greater than 70%. The laser spot completely covers the surface of the composite sensitive structure. The rise time of the laser pulse is less than 10 nanoseconds. The laser power density is calibrated in real time by an optical power meter. The time difference between the laser pulse and the step pressure excitation reaching the surface of the composite sensitive structure is less than 10 nanoseconds.

4. The thermo-coupling and decoupling measurement method based on a composite sensitive structure according to claim 1, characterized in that: During actual measurement, the first mixed signal output by the piezoelectric film in the composite sensitive structure and the second mixed signal output by the thermopile film are simultaneously acquired. The first mixed signal and the second mixed signal are then Fourier transformed to the frequency domain to obtain the first frequency domain mixed signal and the second frequency domain mixed signal. A frequency domain decoupling matrix is ​​constructed based on the dynamic response feature matrix. The frequency domain decoupling matrix is ​​used to perform blind source separation on the first frequency domain mixed signal and the second frequency domain mixed signal, and the frequency domain temperature signal of the frequency domain pressure signal is obtained by solving. The frequency domain pressure signal and the frequency domain temperature signal are respectively subjected to inverse Fourier transform to restore the dynamic pressure time domain waveform and transient temperature time domain change curve that are completely synchronized with the first and second mixed signals in time.

5. The thermo-coupling and decoupling measurement method based on a composite sensitive structure according to claim 1, characterized in that: The method for calculating the thermo-coupling coefficient is as follows: First, under the condition of only step pressure excitation, the first pressure response signal of the piezoelectric film and the second pressure response signal of the thermopile film are obtained respectively. Secondly, under the condition of only pulsed laser thermal excitation, the first thermal response signal of the piezoelectric thin film and the second thermal response signal of the thermopile thin film were obtained respectively. Then, step pressure excitation and pulsed laser thermal excitation are applied simultaneously to obtain the first total response signal and the first thermal response signal of the piezoelectric film, respectively. The difference is used as the thermo-mechanical coupling response quantity of the piezoelectric film. The second total response signal is subtracted from the second pressure response signal and the second thermal response signal, and the difference is used as the thermo-mechanical coupling response quantity of the thermopile film. The thermo-mechanical coupling response quantities of the piezoelectric film and the thermopile film are normalized in the frequency domain to obtain the thermo-mechanical coupling coefficient matrix.

6. The thermo-coupling and decoupling measurement method based on a composite sensitive structure according to claim 5, characterized in that: The process of constructing the frequency domain decoupling matrix is ​​as follows: the amplitude frequency response coefficient and phase frequency response coefficient in the dynamic response characteristic matrix are combined into a complex form of pressure transfer function and heat transfer function, and the thermodynamic coupling coefficient is superimposed on the pressure transfer function and heat transfer function in complex form to form a 2×2 complex transfer function matrix. The frequency domain decoupling matrix is ​​obtained by inverting the complex transfer function matrix at each discrete frequency point.

7. The thermo-mechanical coupling decoupling measurement method based on a composite sensitive structure according to claim 4, characterized in that: The blind source separation process is as follows: the first frequency domain mixed signal and the second frequency domain mixed signal are combined to form a 2×1 frequency domain mixed vector. The frequency domain decoupling matrix is ​​multiplied by the frequency domain mixed vector to obtain a 2×1 frequency domain source vector. The two elements in the frequency domain source vector are the frequency domain pressure signal and the frequency domain temperature signal.

8. The thermo-coupling and decoupling measurement method based on a composite sensitive structure according to claim 1, characterized in that: During the construction of the dynamic response feature matrix, multiple calibration experiments are repeated for different combinations of pressure and thermal excitation amplitudes to obtain multiple sets of dynamic response feature matrices under different amplitude ranges. In actual measurement, the dynamic response feature matrix of the corresponding amplitude range is adaptively selected according to the real-time amplitude of the first mixed signal and the second mixed signal to construct the frequency domain decoupling matrix, so as to eliminate the nonlinear response error introduced by the composite sensitive structure when measuring in a large dynamic range.

9. The thermo-coupling and decoupling measurement method based on a composite sensitive structure according to claim 1, characterized in that: A heat-reflective layer is also provided between the upper surface of the heat insulation layer and the lower surface of the piezoelectric film. The heat-reflective layer is a gold or silver film with a thickness of 100 nanometers to 500 nanometers. It is used to reflect the infrared part that passes through the piezoelectric film during pulsed laser thermal excitation back to the piezoelectric film, so as to improve the absorption efficiency of the piezoelectric film for laser thermal excitation and reduce the direct radiation of heat to the thermopile film.

10. The thermo-coupling and decoupling measurement method based on a composite sensitive structure according to claim 4, characterized in that: The process of restoring the dynamic pressure time-domain waveform and transient temperature time-domain change curve includes a time-domain calibration step: the restored dynamic pressure time-domain waveform and transient temperature time-domain change curve are used as inputs and substituted into the pre-established composite sensitive structure thermo-coupling simulation model. The simulated piezoelectric film output signal and thermopile film output signal are calculated in the forward direction. The simulated output signal is compared with the actual measured first mixed signal and second mixed signal. If the root mean square error between the two exceeds a preset threshold, the frequency domain decoupling matrix is ​​corrected using an iterative deconvolution algorithm, and blind source separation and inverse Fourier transform are performed again until the root mean square error is less than the preset threshold.

Images