A thermal diffusion parameter inversion method and system based on asynchronous sampling spectral features

Abstract

This invention relates to a method and system for inverting thermal diffusion parameters based on asynchronous sampling spectral characteristics, belonging to the field of transient thermal response detection and signal processing technology. The method applies periodic thermal excitation to the sample under test, causing local heat accumulation and subsequent thermal diffusion processes within each excitation cycle. It acquires a measurement signal characterizing the thermal diffusion process; discretely samples the measurement signal, setting the sampling period to be a non-integer multiple of the excitation period to obtain an asynchronous sampling sequence; performs frequency domain transformation on the asynchronous sampling sequence to obtain a discrete spectrum; extracts at least two target discrete spectral lines from the discrete spectrum and constructs feature quantities; and solves for the thermal diffusion characteristic time of the sample under test based on the correspondence between the feature quantities and the thermal response model, ultimately solving for the thermal diffusion parameters of the sample under test. This invention reduces the system's dependence on synchronous triggering accuracy and high-speed sampling hardware, improving parameter extraction efficiency and robustness.

Description

Technical Field

[0001] This invention belongs to the field of transient thermal response detection and signal processing technology, and particularly relates to a method and system for inverting thermal diffusion parameters based on asynchronous sampling spectrum characteristics. Background Technology

[0002] In the thermal characterization of materials and the measurement of transient physical processes, the thermal diffusion process typically corresponds to the establishment, propagation, and decay of the internal temperature field of a sample. Its evolution time constant can reflect the material's thermal transport capacity, local thermal relaxation behavior, and related structural parameters. For thin films, semiconductor wafers, and other micro / nano-scale samples, under periodic pumping excitation, repetitive thermal accumulation and diffusion processes occur in local regions of the sample, causing changes in physical quantities such as refractive index, optical path length, or phase over time. If characteristic time parameters reflecting this process can be accurately extracted from the measurement signal, it can provide important evidence for the analysis of thermal diffusion mechanisms, material performance evaluation, and process status monitoring. However, such thermal responses are usually characterized by strong transientity, short duration, and close coupling with the excitation period, which places high demands on the sampling method, time reference, and signal processing method in the parameter extraction process.

[0003] In existing technologies, common approaches to measuring and analyzing periodic thermal responses include direct high-speed time-domain sampling, reconstructing the transient process through mechanical delay scanning, or performing model fitting after obtaining the response curve. While these methods can estimate the thermal diffusion time constant under certain conditions, they still have significant shortcomings in practical applications: Firstly, if the thermal response time is on the order of microseconds or even shorter timescales, direct time-domain measurement typically requires high sampling bandwidth and strict synchronization triggering conditions, resulting in complex system implementation, large data volume, and sensitivity to electronic link stability. Secondly, while scanning methods can recover the response process point-by-point along the time axis, they suffer from computational redundancy, low measurement efficiency, and are susceptible to mechanical errors, trigger jitter, and long-term drift, hindering rapid and stable engineering applications. For scenarios requiring long-term repeated measurements or dealing with weak thermal response signals, the above methods often struggle to balance time resolution, measurement speed, and system stability.

[0004] To alleviate the pressure of synchronous and high-speed sampling, existing research has attempted to process periodic signals using asynchronous sampling, equivalent time sampling, or frequency domain analysis, aiming to recover high time-resolution information at lower sampling rates. However, existing schemes still have shortcomings when used for solving thermal diffusion parameters. First, many methods still focus on reconstructing an approximately complete time-domain response waveform rather than directly extracting features from the target physical parameters, thus remaining sensitive to the sampling start time, sequence length, and rearrangement accuracy. Second, under the condition of coexisting periodic excitation and asynchronous sampling, the measured signal is usually simultaneously affected by factors such as initial sampling delay, system gain coefficient, and noise disturbance. If fitting is done using only a single spectral feature or a single time-domain curve, the parameter solution is prone to instability, non-uniqueness, or sensitivity to initial values, resulting in insufficient robustness. Third, for pump-probe type thermal diffusion measurements, the sample response often exhibits a first-order or near-first-order decay process driven by periodic thermal excitation. If targeted parameter inversion features cannot be constructed in conjunction with the physical model, it is difficult to stably recover the thermal diffusion characteristic time under limited sampling conditions.

[0005] Based on the above problems, it is necessary to propose a new method and system for inverting thermal diffusion parameters. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for inverting thermal diffusion parameters based on asynchronous sampling spectrum characteristics, so as to solve the problems of low measurement efficiency caused by computational redundancy and poor aluminum rod properties caused by noise disturbance in existing thermal diffusion parameter inversion methods.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows: This invention relates to a method for inverting thermal diffusion parameters based on asynchronous sampling spectral characteristics, comprising the following steps: S1. Apply periodic thermal excitation to the sample to be tested, with an excitation period of... This causes the sample under test to generate local heat accumulation and subsequent heat diffusion processes in each excitation cycle; S2. Acquire measurement signals characterizing the thermal diffusion process; S3. Discretely sample the measured signal and set the sampling period. With incentive cycle If the integer multiple relationship is not satisfied, an asynchronous sampling sequence covering different excitation phase positions is obtained; S4. Perform frequency domain transformation on the asynchronous sampling sequence to obtain the discrete spectrum corresponding to the periodic thermal response; S5. Extract at least two target discrete spectral lines related to thermal diffusion dynamics from the discrete spectrum; S6. Construct a characteristic quantity based on the relationship between the at least two target discrete spectral lines; S7. Preset a thermal response model, and solve the thermal diffusion characteristic time of the sample under test based on the correspondence between the characteristic quantities and the thermal response model. According to the thermal diffusion characteristic time Solve for the thermal diffusion parameters of the sample to be tested.

[0008] In S3, when the measurement signal is discretely sampled, the sampling process is not required to be strictly synchronized with the periodic excitation, nor is the complete transient waveform directly recovered from a single sampling. Instead, the non-integer multiple relationship between the sampling frequency and the excitation frequency is actively utilized to allow the finite rate sampling to gradually sweep across different response phase positions in the long time sequence, thereby encoding the periodic thermal response information into the discrete sampling sequence.

[0009] Preferably, the measurement signal obtained in S2 is one or more of the following: thermally induced phase signal, thermally induced optical path change signal, transmission response signal, reflection response signal, and temperature response signal.

[0010] Preferably, the periodic thermal excitation applied to the sample under test in step S1 is periodic pulsed pump light excitation or periodic modulated pump light excitation. The periodic thermal excitation is applied to a local area of ​​the sample under test to generate a thermal accumulation and thermal diffusion process in the local area, so as to convert the thermal diffusion process into a more sensitive phase response for subsequent processing.

[0011] Preferably, the frequency domain transformation in S4 adopts Fourier transform or equivalent frequency domain analysis, and the obtained discrete spectrum is the amplitude, power or complex spectrum information of each discrete spectral line; the target discrete spectral line in S5 is the discrete spectral line formed in the low-frequency measurable domain after the periodic thermal response is folded or translated under asynchronous sampling conditions. By selecting the target spectral line in the low-frequency measurable domain, the effective transfer and reading of thermal response information under high repetition frequency periodic excitation can be realized.

[0012] Preferably, the feature quantity in S6 is constructed from the feature values ​​of at least two target discrete spectral lines, wherein the feature value is the amplitude, and the feature quantity is constructed by amplitude ratio operation, normalized ratio operation, complex spectrum ratio operation, or weighted combination operation.

[0013] Preferably, in step S7, the thermal diffusion characteristic time of the sample under test is solved based on the correspondence between the characteristic quantity and the thermal response model. The specific steps are as follows: Based on the thermal response model, analytical relationship, or calibration curve, establish the characteristic quantities and thermal diffusion characteristic time. The correspondence between them is used to invert the thermal diffusion characteristic time of the sample under test from the characteristic quantities. .

[0014] Preferably, the S7 preset thermal response model refers to describing the sample response after a single thermal excitation as a first-order or approximately first-order decay dynamic process, and using the thermal diffusion characteristic time... Characterizes the thermal diffusion rate; under periodic thermal excitation conditions, the total response of the sample under test is expressed as the superposition of multiple single thermal responses in each excitation cycle.

[0015] The present invention also relates to a thermal diffusion parameter inversion system based on asynchronous sampling spectral characteristics, comprising: The periodic thermal excitation module is used to apply periodic thermal excitation to the sample under test, with an excitation period of [missing information]. This causes the sample under test to generate local heat accumulation and subsequent heat diffusion processes in each excitation cycle; A response signal acquisition module is used to acquire measurement signals characterizing the thermal diffusion process; An asynchronous sampling module is used to discretely sample the measurement signal and set the sampling period. With incentive cycle If the integer multiple relationship is not satisfied, an asynchronous sampling sequence covering different excitation phase positions is obtained; The frequency domain analysis module is used to perform frequency domain transformation on the asynchronous sampling sequence to obtain the discrete spectrum corresponding to the periodic thermal response. The target spectral line extraction module is used to extract at least two target discrete spectral lines related to thermal diffusion dynamics from the discrete spectrum. A feature relationship construction module is used to construct feature quantities based on the relationship between the at least two target discrete spectral lines; The parameter inversion module is used to preset the thermal response model and solve the thermal diffusion characteristic time of the sample under test based on the correspondence between the characteristic quantities and the thermal response model. According to the thermal diffusion characteristic time Solve for the thermal diffusion parameters of the sample to be tested.

[0016] The asynchronous sampling module consists of a frequency meter, a data acquisition card, an analog-to-digital converter, or other devices capable of sampling the response signal at a preset frequency. The difference between the sampling frequency and the excitation frequency is set to ensure that the sampling points uniformly or quasi-uniformly cover different phase positions within multiple excitation cycles, thereby improving the stability of subsequent feature extraction.

[0017] Preferably, the periodic thermal excitation and response signal acquisition module receives the phase difference output from the synchronous differential heterodyne interferometry phase detection system to convert the refractive index change or optical path change caused by the periodic thermal excitation into a thermally induced phase response signal, thereby acquiring a measurement signal characterizing the thermal diffusion process, and using it as the input signal of the asynchronous sampling module.

[0018] Preferably, the parameter inversion module includes a model establishment and calibration submodule, used to establish the characteristic quantities and thermal diffusion characteristic time. The correspondence between them.

[0019] Compared with the prior art, the technical solution provided by this invention has the following advantages: 1. The thermal diffusion parameter inversion method and system based on asynchronous sampling spectrum characteristics of the present invention, by setting a non-integer multiple relationship between the sampling frequency and the periodic excitation frequency, enables the sampling points to gradually cover different phase positions in multiple excitation cycles, thereby obtaining effective information reflecting the thermal diffusion process without relying on strict synchronous sampling or high-speed continuous time-domain sampling, reducing the system's dependence on synchronous triggering accuracy and high-speed sampling hardware.

[0020] 2. The thermal diffusion parameter inversion method and system based on asynchronous sampling spectrum characteristics involved in this invention does not take the reconstruction of the complete time domain response waveform as the sole objective. Instead, it extracts discrete spectral features through frequency domain transformation and moves the dynamic information in the periodic thermal response to the low-frequency measurable domain for processing, thereby reducing the data processing burden while improving the parameter extraction efficiency. 3. The thermal diffusion parameter inversion method and system based on asynchronous sampling spectral characteristics of the present invention weakens or eliminates the influence of initial sampling delay, system gain coefficient and other common disturbance factors by constructing the relationship between at least two target spectral lines, making the solution of thermal diffusion characteristic time more stable and repeatable, and overcoming the problem that traditional single spectral line or single time domain fitting methods are sensitive to initial conditions.

[0021] 4. The thermal diffusion parameter inversion method and system based on asynchronous sampling spectrum characteristics of the present invention are compatible with various thermal response signal acquisition methods, and are particularly suitable for pump-probe type thermal diffusion measurement and thermal phase response characterization scenarios. It can not only realize the rapid and lossless inversion of thermal diffusion characteristic time, but also has good engineering integration potential and extended application value. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the overall composition of a thermal diffusion parameter inversion system based on asynchronous sampling spectral characteristics; Figure 2 This is a schematic diagram of the periodic thermal excitation and asynchronous sampling principle of the present invention; Figure 3 This is a schematic diagram of the frequency domain characteristics of the asynchronous sampling sequence of the present invention; Figure 4 This is a schematic diagram illustrating the construction of the target spectral line relationship and the principle of mitigating the influence of the initial sampling delay in this invention; Figure 5 This is a schematic diagram illustrating the time-reversal principle of thermal diffusion characteristics in this invention; Figure 6 This is a flowchart of the thermal diffusion parameter inversion method based on asynchronous sampling spectrum characteristics of the present invention; Figure 7 This is a schematic diagram of a platform for inverting thermal diffusion parameters to realize asynchronous sampling spectral characteristics.

[0023] Figure reference numerals: 1-Optical frequency comb path, 2-First half-wave plate (HWP1), 3-First beam splitter (BS1), 4-Mirror, 5-Lens, 6-Second delay line, 7-Acousto-optic modulator (AOM), 8-Second beam splitter (BS2), 9-First slit stop (Slit1), 10-First photodetector (PD1), 11-Second slit stop (Slit2), 12-Second photodetector (PD2), 13-Sample under test, 14-Pump optical path, 15-First delay line, 16-9:1 beam splitter, 17-Third photodetector (PD3), 18-Second half-wave plate (HWP2). Detailed Implementation

[0024] To further understand the content of this invention, the invention will be described in detail with reference to the embodiments. The following embodiments are used to illustrate the invention, but are not intended to limit the scope of the invention.

[0025] This embodiment relates to a method and system for inverting thermal diffusion parameters based on asynchronous sampling spectral characteristics. The system uses periodic thermal excitation as the excitation mode for the sample's thermal response and a response signal acquisition module as the measurement interface for the sample's thermal diffusion process. By maintaining a non-integer multiple relationship between the sampling period and the excitation period, asynchronous sampling sequences covering different excitation phase positions are obtained. Based on this, frequency domain analysis is performed on the asynchronous sampling sequences to extract target discrete spectral lines related to thermal diffusion dynamics. Then, feature quantities insensitive to initial sampling delays are constructed through the relationships between the spectral lines. Finally, by combining the thermal response model, analytical relationships, or calibration curves, a stable inversion of the thermal diffusion characteristic time is achieved.

[0026] The overall composition of the thermal diffusion parameter inversion system based on asynchronous sampling spectrum characteristics is as follows: Figure 1 As shown, it mainly includes a periodic thermal excitation module, a response signal acquisition module, an asynchronous sampling module, a frequency domain analysis module, a target spectral line extraction module, a feature relationship construction module, and a parameter inversion module. Figure 1 The diagram mainly illustrates the connection relationships and signal transmission paths between the various functional modules. The periodic thermal excitation module is used to apply periodic thermal excitation to the sample under test, with an excitation period of [missing information]. The sample under test undergoes localized heat accumulation and subsequent thermal diffusion processes during each excitation cycle, i.e., a periodic thermal response. The response signal acquisition module outputs a measurement signal characterizing this thermal diffusion process. The asynchronous sampling module discretely samples the measurement signal and sets the sampling period. With incentive cycle The sampling sequence does not satisfy an integer multiple relationship to obtain asynchronous sampling sequences covering different excitation phase positions. The frequency domain analysis module performs frequency domain transformation on the asynchronous sampling sequence to obtain a discrete spectrum corresponding to the periodic thermal response. The target spectral line extraction module extracts at least two target spectral lines related to thermal diffusion dynamics from the discrete spectrum. The feature relationship construction module constructs feature quantities based on the target spectral lines, and the parameter inversion module presets a thermal response model and solves for the thermal diffusion characteristic time of the sample under test based on the correspondence between the feature quantities and the thermal response model. Ultimately, based on the characteristic time of thermal diffusion... Solve for the thermal diffusion parameters of the sample to be tested.

[0027] The periodic thermal excitation module can be implemented using periodic pulsed pump light, periodic modulated laser, or other equivalent periodic heat sources. Its function is to create repetitive local thermal disturbances on or near the surface of the sample under test. Preferably, the periodic thermal excitation module is a pump excitation optical path, whose output pump light irradiates a local area of ​​the sample under test to generate local temperature rise, heat accumulation, and radial and depth thermal diffusion processes in the sample. The response signal acquisition module is used to acquire the measurement signal corresponding to the thermal diffusion process. This measurement signal can be a thermally induced phase change signal, a thermally induced optical path change signal, a transmission or reflection intensity change signal, a temperature response signal, or other physical response quantities that can reflect the dynamic process of thermal diffusion. Preferably, it receives the thermally induced phase difference signal output from the front-end phase detection system to improve the detection sensitivity of weak thermal responses.

[0028] like Figure 2 As shown, under periodic thermal excitation, the sample under test will form a single thermal response at the beginning of each excitation cycle, and exhibit a periodic superposition thermal diffusion response in multiple cycles. Figure 2 The upper part shows the periodic thermal excitation sequence, and the lower part shows the corresponding periodic thermal response curve and the distribution of asynchronous sampling points in different excitation periods. In this embodiment, the excitation period of the periodic thermal excitation is assumed to be... The sampling period of the asynchronous sampling module is and make and It does not satisfy the integer multiple relationship, that is, it satisfies , Since the sampling point is an integer, it will not fall on the same phase position repeatedly in each excitation cycle. Instead, it will gradually drift in different cycles as the sampling process progresses, thus forming an asynchronous sampling sequence covering different excitation phase positions. In other words, this embodiment does not require high-speed continuous sampling of the thermal response within a single excitation cycle. Instead, it encodes the information of the sample thermal diffusion process into a discrete sampling sequence by gradually scanning different response phase positions across cycles.

[0029] Furthermore, in this embodiment, the single thermal response is preferably described as a first-order or near-first-order decay process, with the characteristic time constant characterizing the thermal diffusion rate, and this serves as the preset thermal response model; the total response under multiple excitation cycles can be regarded as a periodic superposition of each single thermal response on the time axis. It should be understood that this embodiment is not limited to a single exponential decay model. As long as the established thermal response model can reflect the thermal diffusion dynamics of the sample under periodic thermal excitation conditions and can establish a mapping relationship with the subsequently extracted frequency domain features, it can be used to implement the method of the present invention.

[0030] After the asynchronous sampling module acquires the discrete sampling sequence, the sampling sequence is further input into the frequency domain analysis module. The frequency domain analysis module performs a frequency domain transformation on the asynchronous sampling sequence, preferably using Fourier transform or equivalent frequency domain analysis, to obtain the corresponding discrete spectrum. Due to the sampling period... With incentive cycle If the integer multiple relationship is not satisfied, the information originally contained in the periodic dynamic process in the thermal response of the sample will be represented as several discrete spectra in the discrete spectrum.

[0031] Furthermore, the target spectral line extraction module extracts at least two target discrete spectral lines related to thermal diffusion dynamics from the discrete spectral lines. These target discrete spectral lines are discrete spectral lines formed in the low-frequency measurable domain by folding or translating the periodic thermal response under asynchronous sampling conditions. Figure 3 The diagram illustrates the distribution relationship of the first target discrete spectral line, the second target discrete spectral line, and other discrete spectral lines in the discrete spectrum. This embodiment will use the extraction of two target discrete spectral lines (defined as the first target discrete spectral line and the second target discrete spectral line) as an example for further explanation. In this embodiment, the target spectral line extraction module preferably extracts two target spectral lines from the discrete spectrum that are sensitive to thermal diffusion parameters and suitable for stable extraction. More preferably, the amplitude of the first target spectral line is extracted respectively. With the amplitude of the second target spectral line .

[0032] It should be noted that if only a single target spectral line is used for parameter solving, the spectral line amplitude often includes not only information about the thermal diffusion characteristic time, but is also affected by the initial sampling delay, system scaling factor, and other common perturbation factors, thus making the parameter inversion sensitive to initial conditions. Therefore, this embodiment includes a feature relationship construction module. For example... Figure 4 As shown, the initial sampling delay It will affect the amplitude of the first target spectral line. Second target spectral line amplitude They have a common effect; the feature relationship construction module does not use them directly. or Instead of outputting the result, the characteristic quantity is obtained by constructing the amplitude ratio, normalized ratio, or other spectral relationships between the two. , characteristic quantity The calculation formula is: .because and The common perturbation components contained therein are weakened or eliminated during the relation construction process, thus the resulting characteristic quantities Delay of initial sampling It is less sensitive and more suitable as an input for subsequent thermal diffusion parameter inversion.

[0033] The feature relationship construction module is used to construct feature quantities that are insensitive to non-target factors based on the relationship between the at least two target spectral lines. The basic idea is that if only a single spectral line is used for parameter solving, the spectral line amplitude is usually simultaneously affected by the initial sampling delay, system gain coefficient, and other common perturbation factors, thereby reducing the robustness of parameter inversion. However, by constructing ratio relationships, normalization relationships, or weighted combination relationships between two or more target spectral lines, the influence of the common perturbation factors can be weakened or eliminated, making the final feature quantity mainly related to the thermal diffusion characteristic time. This embodiment uses the relationship between two target spectral lines to construct feature quantities insensitive to non-target factors as an example for illustration. However, extracting three or more discrete spectral lines and performing joint construction can further enhance noise resistance and result robustness. Therefore, regardless of the number of discrete spectral lines extracted, they all fall within the protection scope of this invention.

[0034] It should be noted that in other implementations, amplitude difference normalization, complex spectrum ratios, weighted combination relationships, or other relationship construction methods that can weaken the influence of common perturbation factors can also be used. Essentially, this involves improving the sensitivity of the characteristic quantity to the target thermal diffusion parameter and its robustness to the initial sampling delay by using at least two target spectral lines in combination, rather than by directly fitting a single spectral line.

[0035] like Figure 5 As shown, the obtained feature quantities Further input is made to the parameter inversion module. The parameter inversion module establishes the characteristic quantities based on the thermal response model, analytical relationships, or calibration curves. With characteristic time of thermal diffusion The correspondence between them, thus by The thermal diffusion characteristic time of the sample under test was obtained by inversion. Preferably, the parameter inversion module can be implemented through analytical solution, table lookup solution, fitting solution, or a combination thereof. Analytical solution is suitable for cases where the explicit functional relationship between the response model and the spectral line relationship is known; table lookup solution is suitable for cases where a theoretical derivation, numerical simulation, or experimental calibration has been established beforehand. and The corresponding table case; fitting and solving is suitable for cases where approximate functions or empirical models are used to establish parameter mapping relationships. Regardless of the implementation method, the core relationship is: the model establishment and calibration submodules in the parameter inversion module obtain the feature quantities... By combining the thermal response model or calibration relationship, the characteristic time of thermal diffusion can be obtained through inversion. The model establishment and calibration module can obtain the corresponding relationship through theoretical derivation, numerical simulation, experimental calibration, or a combination thereof, so that the parameter inversion module can achieve interpretable and repeatable parameter solutions under different materials, different structures, or different measurement links.

[0036] Furthermore, the system can be used in conjunction with a phase detection system to convert the refractive index change or optical path change caused by periodic thermal excitation into a phase response, and then perform thermal diffusion parameter inversion through asynchronous sampling and frequency domain analysis. Preferably, the response signal acquisition module receives the phase difference output from the differential heterodyne interferometry phase detection system to characterize the sample thermal response with high sensitivity using a phase-based approach; under these conditions, the present invention can not only achieve rapid solution of thermal diffusion characteristic time, but also perform non-destructive analysis of local thermal processes using phase signals.

[0037] like Figure 6 As shown, this embodiment also provides a method for inverting thermal diffusion parameters based on asynchronous sampling spectral characteristics, the steps of which include: S1. Apply periodic thermal excitation to the sample to be tested, with an excitation period of... This causes the sample under test to generate local heat accumulation and subsequent heat diffusion processes in each excitation cycle; S2. Acquire measurement signals characterizing the thermal diffusion process; S3. Discretely sample the measured signal and set the sampling period. With incentive cycle If the integer multiple relationship is not satisfied, an asynchronous sampling sequence covering different excitation phase positions is obtained; S4. Perform frequency domain transformation on the asynchronous sampling sequence to obtain the discrete spectrum corresponding to the periodic thermal response; S5. Extract at least two target discrete spectral lines related to thermal diffusion dynamics from the discrete spectrum; S6. Construct an initial sampling delay based on the relationship between the at least two target discrete spectral lines. Insensitive feature quantity ; S7. Based on the correspondence between the characteristic quantities and the thermal response model, solve for the thermal diffusion characteristic time of the sample under test. The output results can include not only the thermal diffusion characteristic time It may further include by The derived thermal transport capacity characterization parameters or other relevant thermal parameters are then used. Finally, based on the described thermal diffusion characteristic time... Solve for the thermal diffusion parameters of the sample to be tested. When it is necessary to improve the inversion stability, multiple target spectral lines are further introduced for joint inversion, or the parameters obtained in step 6 are corrected by combining theoretical calibration curves, numerical models and experimental calibration results to obtain more stable and reliable thermal diffusion parameter results.

[0038] In a preferred embodiment of the present invention, the method can be used in conjunction with a front-end phase detection system. For example... Figure 7 As shown, the optical frequency comb outputs the probe light from the optical frequency comb path 1. After being split by the first half-wave plate (HWP1) 2 and the first beam splitter (BS1) 3, the reference light path and the measurement light path are formed respectively. The reference light path is processed by the reflector 4, lens 5, second delay line 6 and acousto-optic modulator (AOM) 7, and then combined with the measurement light path at the second beam splitter (BS2) 8. The measurement light illuminates the sample 13 under test through the lens 5, carrying the information of the sample's thermally induced refractive index change or optical path change back to the detection end. The combined signal passes through the first slit aperture (Slit1) 9 and the second slit aperture (Slit2) 11 respectively and then enters the first photodetector (PD1) 10 and the second photodetector (PD2) 12 for detection. Simultaneously, the 1064 nm laser and its associated pump optical path 14, after being processed by components such as the first delay line 15, the 9:1 beam splitter 16, the third photodetector (PD3) 17, and the second half-wave plate (HWP2) 18, apply periodic thermal excitation to the sample under test 13, thereby generating a thermal accumulation and thermal diffusion process in a local region of the sample. The phase difference signal output by the front-end detection system is further used as the input of the response signal acquisition module of this invention. After asynchronous sampling, frequency domain analysis, spectral line extraction, relation construction, and parameter inversion, the thermal diffusion characteristic time of the sample under test is obtained. .

[0039] It should be noted that, Figure 7 The diagram shown is merely a preferred implementation platform of the present invention. The response signal acquisition module of the present invention is not limited to... Figure 7 The front-end phase detection system shown can also be implemented by other measurement systems capable of outputting characterizations of the thermal diffusion process. For example, in other embodiments, the measurement signal can also be a transmission intensity change signal, a reflection intensity change signal, a thermal radiation response signal, or a response sequence obtained by other optical, electrical, or thermal sensing methods. As long as the response signal can form an extractable target discrete spectral line after asynchronous sampling and frequency domain analysis, and further construct a characteristic quantity for parameter inversion, it falls within the protection scope of this invention.

[0040] The present invention has been described in detail above with reference to the embodiments, but the content described is only a preferred embodiment of the present invention and should not be considered as limiting the scope of the present invention. All equivalent changes and improvements made in accordance with the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A method for inverting thermal diffusion parameters based on asynchronous sampling spectral characteristics, characterized in that, Includes the following steps: S1. Apply periodic thermal excitation to the sample to be tested, with an excitation period of... This causes the sample under test to generate local heat accumulation and subsequent heat diffusion processes in each excitation cycle; S2. Acquire measurement signals characterizing the thermal diffusion process; S3. Discretely sample the measured signal and set the sampling period. With incentive cycle If the integer multiple relationship is not satisfied, an asynchronous sampling sequence covering different excitation phase positions is obtained; S4. Perform frequency domain transformation on the asynchronous sampling sequence to obtain the discrete spectrum corresponding to the periodic thermal response; S5. Extract at least two target discrete spectral lines related to thermal diffusion dynamics from the discrete spectrum; S6. Construct a characteristic quantity based on the relationship between the at least two target discrete spectral lines; S7. Preset a thermal response model, and solve the thermal diffusion characteristic time of the sample under test based on the correspondence between the characteristic quantities and the thermal response model. According to the thermal diffusion characteristic time Solve for the thermal diffusion parameters of the sample to be tested.

2. The thermal diffusion parameter inversion method based on asynchronous sampling spectral characteristics according to claim 1, characterized in that: The measurement signal obtained in S2 is one or more of the following: thermally induced phase signal, thermally induced optical path change signal, transmission response signal, reflection response signal, and temperature response signal.

3. The thermal diffusion parameter inversion method based on asynchronous sampling spectral characteristics according to claim 1, characterized in that: The periodic thermal excitation applied to the sample under test by S1 is either periodic pulsed pump light excitation or periodic modulated pump light excitation. The periodic thermal excitation is applied to a local area of ​​the sample under test to generate a heat accumulation and heat diffusion process in the local area.

4. The thermal diffusion parameter inversion method based on asynchronous sampling spectral characteristics according to claim 1, characterized in that: The frequency domain transformation in S4 uses Fourier transform or equivalent frequency domain analysis to obtain discrete spectrum, which is the amplitude, power or complex spectrum information of each discrete spectral line; the target discrete spectral line in S5 is a discrete spectral line formed in the low-frequency measurable domain after the periodic thermal response is folded or translated under asynchronous sampling conditions.

5. The thermal diffusion parameter inversion method based on asynchronous sampling spectral characteristics according to claim 1, characterized in that: The feature quantity in S6 is constructed from the feature values ​​of at least two target discrete spectral lines. The feature value is the amplitude. The feature quantity is constructed by amplitude ratio operation, normalized ratio operation, complex spectrum ratio operation or weighted combination operation.

6. The thermal diffusion parameter inversion method based on asynchronous sampling spectral characteristics according to claim 5, characterized in that: In step S7, the thermal diffusion characteristic time of the sample under test is calculated based on the correspondence between the characteristic quantity and the thermal response model. The specific steps are as follows: Based on the thermal response model, analytical relationship, or calibration curve, establish the characteristic quantities and thermal diffusion characteristic time. The correspondence between them is used to invert the thermal diffusion characteristic time of the sample under test from the characteristic quantities. .

7. The thermal diffusion parameter inversion method based on asynchronous sampling spectral characteristics according to claim 1, characterized in that: The S7 preset thermal response model refers to describing the sample response after a single thermal excitation as a first-order or approximately first-order decay dynamic process, and using the thermal diffusion characteristic time as the basis. Characterizes the thermal diffusion rate; under periodic thermal excitation conditions, the total response of the sample under test is expressed as the superposition of multiple single thermal responses in each excitation cycle.

8. A thermal diffusion parameter inversion system based on asynchronous sampling spectral characteristics, characterized in that, It includes: The periodic thermal excitation module is used to apply periodic thermal excitation to the sample under test, with an excitation period of [missing information]. This causes the sample under test to generate local heat accumulation and subsequent heat diffusion processes in each excitation cycle; A response signal acquisition module is used to acquire measurement signals characterizing the thermal diffusion process; An asynchronous sampling module is used to discretely sample the measurement signal and set the sampling period. With incentive cycle If the integer multiple relationship is not satisfied, an asynchronous sampling sequence covering different excitation phase positions is obtained; The frequency domain analysis module is used to perform frequency domain transformation on the asynchronous sampling sequence to obtain the discrete spectrum corresponding to the periodic thermal response. The target spectral line extraction module is used to extract at least two target discrete spectral lines related to thermal diffusion dynamics from the discrete spectrum. A feature relationship construction module is used to construct feature quantities based on the relationship between the at least two target discrete spectral lines; The parameter inversion module is used to preset the thermal response model and solve the thermal diffusion characteristic time of the sample under test based on the correspondence between the characteristic quantities and the thermal response model. According to the thermal diffusion characteristic time Solve for the thermal diffusion parameters of the sample to be tested.

9. The thermal diffusion parameter inversion system based on asynchronous sampling spectral characteristics according to claim 8, characterized in that: The periodic thermal excitation and response signal acquisition module receives the phase difference output from the synchronous differential heterodyne interferometry phase detection system to convert the refractive index change or optical path change caused by the periodic thermal excitation into a thermally induced phase response signal, thereby acquiring a measurement signal characterizing the thermal diffusion process and using it as the input signal of the asynchronous sampling module.

10. The thermal diffusion parameter inversion system based on asynchronous sampling spectral characteristics according to claim 8, characterized in that: The parameter inversion module includes a model building and calibration submodule, used to establish the characteristic quantities and thermal diffusion characteristic time. The correspondence between them.

Images