CN114427838A - Method and system for predicting and evaluating thickness of medium based on reflection terahertz spectrum - Google Patents
Method and system for predicting and evaluating thickness of medium based on reflection terahertz spectrum Download PDFInfo
- Publication number
- CN114427838A CN114427838A CN202210023349.1A CN202210023349A CN114427838A CN 114427838 A CN114427838 A CN 114427838A CN 202210023349 A CN202210023349 A CN 202210023349A CN 114427838 A CN114427838 A CN 114427838A
- Authority
- CN
- China
- Prior art keywords
- thickness
- terahertz
- reflection
- region
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/06—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material
- G01B11/0616—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material of coating
- G01B11/0625—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material of coating with measurement of absorption or reflection
Abstract
The invention provides a method and a system for predicting and evaluating medium thickness based on a reflection terahertz spectrum, wherein the prediction method comprises the following steps: acquiring a terahertz reflection signal of a region to be detected and preprocessing the terahertz reflection signal; inputting the preprocessed terahertz reflection signal of the region to be detected into a thickness prediction model to obtain a thickness value of the region to be detected output by the thickness prediction model. The evaluation method comprises the following steps: and comparing the thickness prediction result with a thickness result obtained by utilizing the time delay principle to further verify the accuracy and the applicability of the prediction model. The prediction method provided by the invention only needs to consider the reference waveform and the known reference thickness and establish a fitting model of the reference waveform and the known reference thickness, thereby avoiding the thickness calculation error caused by signal mixing due to the thinner medium thickness and waveform broadening due to terahertz wave transmission dispersion, providing a new idea for the thickness measurement of an optical thin sample and a thick sample, and providing a new direction for the thickness nondestructive detection of industrial parts with complex structures.
Description
Technical Field
The invention relates to the technical field of terahertz intelligent detection, in particular to a method and a system for predicting and evaluating medium thickness based on a reflection terahertz spectrum.
Background
The composite material has wide application in industrial production, wherein other media are bonded on the surface of a metal or metalloid substrate to play a beneficial role in protecting or enhancing the performance, and the service performance and the service life of key parts are improved. For example, the aero-engine utilizes a specific coating to realize the effects of oxidation resistance, heat insulation, wear resistance and the like, and the service life of the engine is prolonged; exterior anti-ablation coatings for high speed aircraft, launch vehicles, and the like; the surface of the automobile shell is coated with a film to achieve the characteristics of corrosion resistance and wear resistance, and paint films with different colors are attractive; coating structure during chip processing, etc. Due to the influence of factors such as a preparation process, a workpiece structure, application complexity and the like, a series of new quality detection and reliability evaluation problems are generated, and the quality analysis and evaluation of the protective media of the parts and equipment by applying a nondestructive testing technology are very important at present.
The thickness of the dielectric layer is an important index for evaluating the reliability and application safety of the workpiece preparation process, and has important significance for improving the performance and the service efficiency of the workpiece, and the existing nondestructive testing technology system is continuously developed to deal with the problems. The conventional medium layer thickness measuring methods include ultrasonic thickness measurement, X-ray thickness measurement, acoustic emission thickness measurement, eddy current thickness measurement and the like, and have certain limitations, the ultrasonic thickness measurement requires that the surface of a test piece is smooth and the detection precision is not high, X-rays possibly influence detection personnel, the acoustic emission method needs to apply load for detection, dynamic monitoring is achieved, and the eddy current thickness measurement is greatly influenced by the conductivity and the size of a material.
In recent years, terahertz technology is continuously applied to the field of intelligent detection as a new monitoring method. As a new radiation source, terahertz waves have the characteristics of low single photon energy, strong reflectivity to metal and metalloid materials, high penetrability to most nonmetal materials and the like, the terahertz detection technology has shown great application potential in the fields of aerospace, cultural relic detection, energy power, biomedicine and the like, and particularly is an ideal optimization scheme for quality detection of various novel composite materials.
Two common terahertz thickness measuring methods are available, one method is to extract the refractive index of a standard sample with known thickness by using a transmission mode, and then measure the actual sample thickness by using a reflection mode and combining a time delay principle. The other method is to measure the thickness of a sample in a reflection mode, extract the refractive index of the sample by taking a reflector as a reference signal, establish an interaction model of a substance and terahertz waves, compare a theoretical reflection time domain signal with an experimental measurement signal, and continuously iterate by means of an optimization algorithm to further determine the thickness of the sample.
Disclosure of Invention
The invention provides a method and a system for predicting and evaluating the thickness of a medium based on a reflection terahertz spectrum, which are used for improving the defect of detection precision when the existing terahertz thickness measurement method is used for detecting a medium of a bonding protection layer on a metal or metalloid substrate, and realizing the accurate measurement of the thickness of the medium.
The invention provides a medium thickness prediction method based on a reflection terahertz spectrum, which comprises the following steps:
acquiring a terahertz reflection signal of a region to be detected and preprocessing the terahertz reflection signal;
inputting the preprocessed terahertz reflection signal of the region to be detected into a thickness prediction model to obtain a thickness prediction value of the region to be detected, which is output by the thickness prediction model;
the thickness prediction model is obtained by training through a partial least squares regression method based on the terahertz reflection signals of the known thickness reference regions and the thickness values of the known thickness reference regions.
According to the medium thickness prediction method based on the reflection terahertz spectrum, provided by the invention, the terahertz reflection signal of the region to be measured is obtained by vertically irradiating the region to be measured with terahertz pulses.
According to the medium thickness prediction method based on the reflection terahertz spectrum, provided by the invention, the terahertz reflection signal of the region to be detected comprises three complete reflection echoes.
According to the medium thickness prediction method based on the reflection terahertz spectrum, provided by the invention, the pretreatment comprises the following steps: baseline correction and wavelet filtering.
According to the medium thickness prediction method based on the reflection terahertz spectrum, provided by the invention, the thickness prediction model is trained by a partial least square regression method and comprises the following steps:
obtaining an independent variable matrix according to the terahertz reflection signals of the known thickness reference region, obtaining a dependent variable vector according to the thickness values of the known thickness reference region, and normalizing the independent variable matrix and the dependent variable vector to obtain a first normalized matrix and a second normalized matrix;
and establishing a regression equation by adopting a partial least squares regression method according to the first standardized matrix and the second standardized matrix, and training to obtain a thickness prediction model taking the spectrum as an independent variable and the thickness value as a dependent variable.
The invention also provides an evaluation method, which comprises the following steps:
obtaining a thickness measurement value of a region to be measured according to a time delay principle of the terahertz time-domain signal;
obtaining a thickness prediction value of the area to be measured by adopting the medium thickness prediction method based on the reflection terahertz spectrum;
and obtaining the evaluation of the prediction accuracy according to the thickness measurement value and the thickness prediction value of the area to be measured.
According to the evaluation method provided by the invention, the obtaining of the thickness measurement value of the region to be measured according to the time delay principle of the terahertz time-domain signal comprises the following steps:
and separating a first reflection signal and a second reflection signal of the terahertz time-domain signal of the region to be measured, and obtaining a thickness measurement value of the region to be measured according to the time interval between the peak value of the first reflection signal and the peak value of the second reflection signal and the refractive index of the medium.
The invention also provides a medium thickness prediction system based on the reflection terahertz spectrum, which comprises the following components:
the signal acquisition module is used for acquiring terahertz reflection signals of a region to be detected and preprocessing the terahertz reflection signals;
the thickness prediction module is used for inputting the preprocessed terahertz reflection signals of the region to be detected into a thickness prediction model to obtain a thickness prediction value of the region to be detected, which is output by the thickness prediction model;
the thickness prediction model is obtained by training through a partial least squares regression method based on the terahertz reflection signals of the known thickness reference regions and the thickness values of the known thickness reference regions.
The present invention also provides an evaluation system comprising:
the measuring module is used for obtaining a thickness measuring value of the area to be measured according to a time delay principle of the terahertz time-domain signal;
the prediction module is used for obtaining a thickness prediction value of the region to be measured by adopting the medium thickness prediction method based on the reflection terahertz spectrum;
and the evaluation module is used for obtaining the evaluation of the prediction accuracy according to the thickness measurement value and the thickness prediction value of the area to be measured.
The method and the system for predicting and evaluating the thickness of the medium based on the reflection terahertz spectrum have high practicability and avoid the difficulty in manufacturing a standard transmission test piece aiming at the nondestructive detection of the surface medium with a complex surface structure and application taking metal or metalloid as a substrate.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a thickness prediction method based on a reflection terahertz spectrum according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a reflection measurement model provided by an embodiment of the invention;
FIG. 3 is a schematic diagram of a spectrum of a reflected signal provided by an embodiment of the present invention;
FIG. 4 is a diagram of a terahertz reflection spectrum collected according to an embodiment of the present invention;
FIG. 5 is a refractive index spectrum of a thermal barrier coating provided by an embodiment of the present invention;
FIG. 6 is a graph comparing the predicted regression thickness at different positions of the blade with the terahertz thickness measurement result according to the embodiment of the present invention;
FIG. 7 is a thickness difference between the predicted regression thickness at different positions of the blade and the terahertz thickness measurement result provided by the embodiment of the invention;
FIG. 8 is a schematic structural diagram of a thickness prediction system based on reflected terahertz spectroscopy according to an embodiment of the present invention;
FIG. 9 is a schematic flow chart of an evaluation method provided by an embodiment of the present invention;
FIG. 10 is a second schematic flow chart of an evaluation method according to an embodiment of the present invention;
fig. 11 is a schematic structural diagram of an evaluation system according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The thickness prediction method based on the reflection terahertz spectrum of the present invention is described below with reference to fig. 1 to 7, and in this embodiment, is actually a thickness prediction method for a strong reflection substrate quasi-single layer medium, as shown in fig. 1, the method includes the following steps:
it should be noted that the preprocessing can reduce the signal peak phase extraction error caused by oscillation and noise.
102, inputting the preprocessed terahertz reflection signals of the region to be detected into a thickness prediction model to obtain a thickness prediction value of the region to be detected, which is output by the thickness prediction model;
the thickness prediction model is obtained by training through a partial least squares regression method based on the terahertz reflection signals of the known thickness reference regions and the thickness values of the known thickness reference regions.
It should be noted that the region to be measured is generally any region on the metal or metalloid substrate protective layer medium with a complex structure, and the terahertz reflection signal includes an upper surface signal and a lower surface signal of the medium layer obtained during the reflection terahertz time-domain spectroscopy measurement.
According to the thickness prediction method based on the reflection terahertz spectrum, provided by the embodiment of the invention, the thickness of any region of the dielectric layer is predicted by establishing and training the thickness prediction model, the method is closer to the actual detection requirement, and a new direction is provided for the thickness nondestructive detection of industrial parts with complex structures.
In at least one embodiment of the present invention, the signal preprocessing includes baseline correction and wavelet filtering of the signal to mitigate signal peak phase extraction errors due to ringing and noise.
The pretreatment result should satisfy: the signal has no obvious distortion and deformation phenomenon; the root mean square error of the signals before and after preprocessing is small, and the total energy ratio of the signals before and after preprocessing is generally more than 99 percent.
Fig. 2 is a schematic diagram of a reflection measurement model according to an embodiment of the present invention, where E0 is an incident terahertz pulse, E1 is a reflected signal of an upper surface of a sample to be measured, E2 is a reflected signal of a lower surface of the sample, and E3 is a reflected signal between the sample and a metal base layer.
In at least one embodiment of the invention, the terahertz reflection signal of the region to be measured is obtained by vertically irradiating a terahertz pulse to the region to be measured.
In at least one embodiment of the invention, the terahertz reflection signal of the region to be measured includes three complete reflection echoes, so that the signal integrity of the second reflection echo can be ensured.
In at least one embodiment of the present invention, the thickness prediction model trained by partial least squares regression comprises:
obtaining an independent variable matrix according to the terahertz reflection signals of the known thickness reference region, obtaining a dependent variable vector according to the thickness value of the known thickness reference region, and carrying out standardization processing on the independent variable matrix and the dependent variable vector to obtain a first standardization matrix and a second standardization matrix;
and establishing a regression equation by adopting a partial least squares regression method according to the first standardized matrix and the second standardized matrix, and training to obtain a thickness prediction model taking the spectrum as an independent variable and the thickness value as a dependent variable.
It should be noted that the specific modeling process of partial least squares regression is as follows:
firstly, taking a terahertz reflection spectrum data array of a reference region with known thickness as an independent variable matrix X, taking a thickness value at a corresponding position as a dependent variable vector Y, and standardizing the X and the Y to obtain a first standardized matrix and a second standardized matrix E0 and F0;
keeping t1 as the 1 st spectrum principal component of E0, u1 as the first thickness principal component of F0, requiring the maximum correlation of t1 and u1, and respectively solving the regression equations of E0 and F0 to t 1;
thirdly, residual matrixes E1 and F1 are used for replacing E0 and F0, a spectrum principal component t2 and a thickness principal component u2 are solved, and regression equations of the E1 and the F1 to the t2 are solved respectively;
and fourthly, determining the number of the main components by using cross validation to obtain regression equations of the E0 and the F0 about the r main components, and obtaining the regression equations about the spectral data and the thickness values.
The partial least square thickness prediction model provided by the embodiment of the invention only needs to consider the reference waveform and the known reference thickness and establish the fitting model of the reference waveform and the known reference thickness, so that the extraction error caused by waveform broadening, deformation and the like in time domain calculation is avoided, the error caused by phase shift in the process of taking a mirror surface as a reference is also avoided, the partial least square thickness prediction model is closer to the actual detection requirement, and a new direction is provided for the thickness nondestructive detection of industrial parts with complex structures.
The evaluation method provided by the invention is described below, the embodiment of the invention takes the turbine blade as an embodiment, the thickness of the thermal barrier coating in different areas of the blade is predicted by using the method, and the prediction result is compared with the time domain calculation result to illustrate the correctness and applicability of the method.
As shown in fig. 9, an evaluation method according to an embodiment of the present invention includes:
and 203, obtaining the evaluation of prediction accuracy according to the thickness measurement value and the thickness prediction value of the area to be measured.
In at least one embodiment of the present invention, the obtaining a thickness measurement value of a region to be measured according to a time delay principle of a terahertz time-domain signal includes:
and separating a first reflection signal and a second reflection signal of the terahertz time-domain signal of the region to be measured, and obtaining a thickness measurement value of the region to be measured according to the time interval between the peak value of the first reflection signal and the peak value of the second reflection signal and the refractive index of the medium.
It should be noted that, with the time interval Δ t between the peak value of the first reflection E1 and the peak value of the second reflection E2 of the time domain signal of the detection area with unknown thickness, the dielectric layer thickness d is calculated by combining the formula d ═ c ×. Δ t/(2n) under the condition that the refractive index n of the dielectric is known, where c ═ 3 ×.10 ×8m/s。
In at least one embodiment of the present invention, the method for calculating the refractive index of the medium includes:
obtaining respective frequency domain spectrum E by Fourier transform1(omega) and E2(ω), the reflection transfer function is then equation 2:
wherein, tas、tsa、rsm、rasThe terahertz wave is obtained by utilizing a Fresnel formula, namely the transmission coefficient of the terahertz wave transmitted from the air to the inside of the medium layer, the transmission coefficient transmitted from the inside of the medium layer to the air layer, the reflection coefficient between the medium layer and the metal base interface, and the reflection coefficient reflected back to the air on the upper surface of the medium layer, d is the thickness of a sample at the position, omega is angular frequency, rho (omega) is amplitude characteristic, phi (omega) is phase characteristic, and the complex refractive index of the sample is
The refractive index expression of the obtained sample is shown as the formula1 is as follows:
in the refractive index spectrum of the thermal barrier coating shown in FIG. 5, the average refractive index of the medium is 4.10 in the frequency range of 0.3-1.2 THz.
In at least one embodiment of the present invention, the evaluation method specifically includes the following steps:
step a: acquiring a time domain reflection signal vertically irradiating a reference region with known thickness of the turbine blade by using a reflection terahertz time domain spectroscopy system, as shown in FIG. 4;
step b: medium upper surface reflection signal E separating time domain reflection signals of known thickness reference regions1(t) and lower surface reflection signal E2(t), then calculating a blade refractive index spectrum, and calculating the average refractive index of the medium according to the blade refractive index spectrum, wherein the result is shown in FIG. 5;
step c: measuring a time domain reflection signal of an unknown thickness region of a test piece by using a terahertz system, taking a time domain reflection spectrum of a sample with a known thickness region as an independent variable and taking a corresponding thickness value as a dependent variable, establishing a thickness prediction model of the thickness and the spectrum by using a partial least squares regression method, and inputting time domain reflection signals of other regions into a trained thickness prediction model to obtain a corresponding predicted thickness;
step d: as the refractive indexes of the medium of the protective layer of the test piece are the same at all positions, the thickness of the medium layer in the unknown thickness area is calculated by combining the time delay principle of the terahertz signal on the premise that the average refractive index of the known medium is 4.10, the prediction result is compared with the terahertz thickness calculation result, and the accuracy of the prediction result is verified.
Fig. 6 and 7 are a regression thickness prediction result, a terahertz thickness measurement result and a thickness difference between the two results at different positions of the blade provided by the embodiment of the invention, and it can be seen from the graphs that the method of the embodiment of the invention is used for predicting the thicknesses of 18 different positions in other regions of the blade, and the similarity coefficients R of the two results296.81% is achieved, and the difference in thickness between the two results is less than 3um, therefore, the thickness prediction and the calculation result of the invention have consistency, and the prediction precision of the invention is effectively proved.
In at least one embodiment of the present invention, as shown in fig. 10, the evaluation method includes:
and step 405, comparing the thickness prediction result with a thickness result obtained by the terahertz time-domain method, and calculating the absolute error and the correlation coefficient of the two thickness results.
The thickness prediction system based on the reflection terahertz spectrum provided by the invention is described below, the thickness prediction system based on the reflection terahertz spectrum described below and the thickness prediction method based on the reflection terahertz spectrum described above can be correspondingly referred to each other, and the system is mainly applied to the thickness prediction of the quasi-single-layer medium.
As shown in fig. 8, an embodiment of the present invention provides a system for predicting a thickness of a single-layer medium based on a reflected terahertz spectrum, including:
the signal acquisition module 801 is used for acquiring terahertz reflection signals of a region to be detected and preprocessing the terahertz reflection signals;
the thickness prediction module 802 is configured to input the preprocessed terahertz reflection signal of the region to be detected into a thickness prediction model, and obtain a thickness prediction value of the region to be detected output by the thickness prediction model;
the thickness prediction model is obtained by training through a partial least square regression method based on terahertz reflection signals of known thickness reference regions and thickness values of the known thickness reference regions.
According to the thickness prediction method based on the reflection terahertz spectrum, provided by the embodiment of the invention, the thickness of any region of the dielectric layer is predicted by establishing and training the thickness prediction model, the method is closer to the actual detection requirement, and a new direction is provided for the thickness nondestructive detection of industrial parts with complex structures.
In at least one embodiment of the invention, in the system, the terahertz reflected signal of the region to be measured is obtained by vertically irradiating the region to be measured with a terahertz pulse.
In at least one embodiment of the present invention, in the system, the terahertz reflected signal of the region to be measured includes at least three complete reflected echoes.
In at least one embodiment of the invention, the system wherein the pre-processing comprises: baseline correction, SG smoothing filtering, and wavelet transform.
In at least one embodiment of the present invention, the system wherein the training of the thickness prediction model by partial least squares regression comprises:
obtaining an independent variable matrix according to the terahertz reflection signals of the known thickness reference region, obtaining a dependent variable vector according to the thickness values of the known thickness reference region, and normalizing the independent variable matrix and the dependent variable vector to obtain a first normalized matrix and a second normalized matrix;
and establishing a regression equation by adopting a partial least squares regression method according to the first standardized matrix and the second standardized matrix, and training to obtain a thickness prediction model taking the spectrum as an independent variable and the thickness value as a dependent variable.
In at least one embodiment of the invention, the acquisition module is used for acquiring the terahertz reflection signal vertically irradiated to the region to be measured and the terahertz reflection signal vertically irradiated to the standard thickness reference region.
The evaluation system provided by the invention is described below, the evaluation system described below and the evaluation method described above can be correspondingly referred to each other, and the system is mainly applied to the accuracy evaluation of the thickness prediction method of the quasi-single-layer medium.
As shown in fig. 11, an embodiment of the present invention provides an evaluation system, including:
the measuring module 301 obtains a thickness measurement value of the region to be measured according to a time delay principle of the terahertz time-domain signal;
the prediction module 302 is used for obtaining a thickness prediction value of the region to be measured by adopting the medium thickness prediction method based on the reflection terahertz spectrum;
and the evaluation module 303 is used for obtaining the evaluation of the prediction accuracy according to the thickness measurement value and the thickness prediction value of the area to be measured.
The evaluation system provided by the embodiment of the invention is used for comparing the calculated measured value with the predicted value result, and explaining the accuracy and the applicability of the prediction method.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A medium thickness prediction method based on a reflection terahertz spectrum is characterized by comprising the following steps:
acquiring a terahertz reflection signal of a region to be detected and preprocessing the terahertz reflection signal;
inputting the preprocessed terahertz reflection signal of the region to be detected into a thickness prediction model to obtain a thickness prediction value of the region to be detected, which is output by the thickness prediction model;
the thickness prediction model is obtained by training through a partial least squares regression method based on the terahertz reflection signals of the known thickness reference regions and the thickness values of the known thickness reference regions.
2. The method for predicting the thickness of a medium based on the reflection terahertz spectrum according to claim 1, wherein the terahertz reflection signal of the region to be measured is obtained by vertically irradiating a terahertz pulse to the region to be measured.
3. The method for predicting the thickness of the medium based on the reflected terahertz spectrum according to claim 2, wherein the terahertz reflected signal of the region to be measured comprises three complete reflected echoes, and the three complete reflected echoes are used for ensuring the integrity of the reflected signals on the upper surface and the lower surface of the medium when the reflected signal is separated by selecting a time window.
4. The method for predicting the thickness of the medium based on the reflection terahertz spectrum according to any one of claims 1 to 3, wherein the preprocessing comprises: baseline correction and wavelet filtering, the preprocessing being used to attenuate signal peak phase extraction errors due to ringing and noise.
5. The method for predicting the thickness of the medium based on the reflection terahertz spectrum according to any one of claims 1 to 3, wherein the thickness prediction model is trained by a partial least squares regression method and comprises the following steps:
obtaining an independent variable matrix according to the terahertz reflection signals of the known thickness reference region, obtaining a dependent variable vector according to the thickness values of the known thickness reference region, and normalizing the independent variable matrix and the dependent variable vector to obtain a first normalized matrix and a second normalized matrix;
and establishing a regression equation by adopting a partial least squares regression method according to the first standardized matrix and the second standardized matrix, and training to obtain a thickness prediction model taking the spectrum as an independent variable and the thickness value as a dependent variable.
6. An evaluation method, characterized in that the method comprises:
obtaining a thickness measurement value of the region to be measured according to a time delay principle of the terahertz time-domain signal;
obtaining a predicted thickness value of the region to be measured by adopting the medium thickness prediction method based on the reflection terahertz spectrum of any one of claims 1 to 5;
and obtaining the evaluation of the prediction accuracy according to the thickness measurement value and the thickness prediction value of the area to be measured.
7. The evaluation method according to claim 6, wherein the obtaining of the thickness measurement value of the region to be measured according to the time delay principle of the terahertz time-domain signal comprises:
and separating a first reflection signal and a second reflection signal of the terahertz time-domain signal of the region to be measured, and obtaining a thickness measurement value of the region to be measured according to the time interval between the peak value of the first reflection signal and the peak value of the second reflection signal and the refractive index of the medium.
8. The evaluation method according to claim 7, wherein the medium refractive index is obtained using formula 1:
wherein n iss(ω) represents the refractive index of the medium, tas、tsa、rsm、rasThe terahertz wave obtained by utilizing a Fresnel formula respectively represents a transmission coefficient of the terahertz wave transmitted from the air to the inside of the dielectric layer, a transmission coefficient transmitted from the inside of the dielectric layer to the air layer, a reflection coefficient between the dielectric layer and a metal substrate interface and a reflection coefficient reflected from the surface of the dielectric layer back to the air, d is the thickness of the dielectric layer, omega is the angular frequency, and phi (omega) is the phase characteristic.
9. A medium thickness prediction system based on a reflection terahertz spectrum is characterized by comprising:
the signal acquisition module is used for acquiring terahertz reflection signals of a region to be detected and preprocessing the terahertz reflection signals;
the thickness prediction module is used for inputting the preprocessed terahertz reflection signals of the region to be detected into a thickness prediction model to obtain a thickness prediction value of the region to be detected, which is output by the thickness prediction model;
the thickness prediction model is obtained by training through a partial least squares regression method based on the terahertz reflection signals of the known thickness reference regions and the thickness values of the known thickness reference regions.
10. An evaluation system, comprising:
the measuring module is used for obtaining a thickness measuring value of the area to be measured according to a time delay principle of the terahertz time-domain signal;
the prediction module is used for obtaining a predicted thickness value of the area to be measured by adopting the medium thickness prediction method based on the reflection terahertz spectrum as claimed in any one of claims 1 to 5;
and the evaluation module is used for obtaining the evaluation of the prediction accuracy according to the thickness measurement value and the thickness prediction value of the area to be measured.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210023349.1A CN114427838A (en) | 2022-01-10 | 2022-01-10 | Method and system for predicting and evaluating thickness of medium based on reflection terahertz spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210023349.1A CN114427838A (en) | 2022-01-10 | 2022-01-10 | Method and system for predicting and evaluating thickness of medium based on reflection terahertz spectrum |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114427838A true CN114427838A (en) | 2022-05-03 |
Family
ID=81311442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210023349.1A Pending CN114427838A (en) | 2022-01-10 | 2022-01-10 | Method and system for predicting and evaluating thickness of medium based on reflection terahertz spectrum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114427838A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114894105A (en) * | 2022-05-16 | 2022-08-12 | 西南科技大学 | Method and system for measuring thickness of non-metallic material in atmospheric environment |
| CN115290597A (en) * | 2022-10-08 | 2022-11-04 | 首都师范大学 | Terahertz technology-based method and system for detecting coating adhesion-free defect |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101750024A (en) * | 2009-12-15 | 2010-06-23 | 山西太钢不锈钢股份有限公司 | Near infrared spectrum detection method for insulation coating layer thickness of silicon steel |
| CN104169677A (en) * | 2012-02-08 | 2014-11-26 | 霍尼韦尔阿斯卡公司 | Caliper coating measurement on continuous non-uniform web using THz sensor |
| CN104807496A (en) * | 2014-01-28 | 2015-07-29 | Abb技术有限公司 | Sensor System And Method For Characterizing A Coated Body |
| CN106596462A (en) * | 2016-12-19 | 2017-04-26 | 中国矿业大学 | Paper sheet quantitative detection method based on terahertz phase shifting characteristics and particle swarm algorithm |
| CN108519059A (en) * | 2018-04-20 | 2018-09-11 | 中国矿业大学 | Thermal barrier coating multi-layered thickness detection method based on reflection-type terahertz time-domain spectroscopic technology |
| US20190078873A1 (en) * | 2016-04-04 | 2019-03-14 | Tetechs Inc. | Methods and systems for thickness measurement of multi-layer structures |
| CN110081826A (en) * | 2019-04-29 | 2019-08-02 | 中国矿业大学 | Heat-barrier coating ceramic layer thickness measure new method based on Terahertz Technology |
| US20200116473A1 (en) * | 2017-03-30 | 2020-04-16 | Iucf-Hyu (Industry-University Cooperation Foundation Hanyang University) | Thickness measurement apparatus, thickness measurement method, and thickness measurement program |
-
2022
- 2022-01-10 CN CN202210023349.1A patent/CN114427838A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101750024A (en) * | 2009-12-15 | 2010-06-23 | 山西太钢不锈钢股份有限公司 | Near infrared spectrum detection method for insulation coating layer thickness of silicon steel |
| CN104169677A (en) * | 2012-02-08 | 2014-11-26 | 霍尼韦尔阿斯卡公司 | Caliper coating measurement on continuous non-uniform web using THz sensor |
| CN104807496A (en) * | 2014-01-28 | 2015-07-29 | Abb技术有限公司 | Sensor System And Method For Characterizing A Coated Body |
| US20190078873A1 (en) * | 2016-04-04 | 2019-03-14 | Tetechs Inc. | Methods and systems for thickness measurement of multi-layer structures |
| CN106596462A (en) * | 2016-12-19 | 2017-04-26 | 中国矿业大学 | Paper sheet quantitative detection method based on terahertz phase shifting characteristics and particle swarm algorithm |
| US20200116473A1 (en) * | 2017-03-30 | 2020-04-16 | Iucf-Hyu (Industry-University Cooperation Foundation Hanyang University) | Thickness measurement apparatus, thickness measurement method, and thickness measurement program |
| CN108519059A (en) * | 2018-04-20 | 2018-09-11 | 中国矿业大学 | Thermal barrier coating multi-layered thickness detection method based on reflection-type terahertz time-domain spectroscopic technology |
| CN110081826A (en) * | 2019-04-29 | 2019-08-02 | 中国矿业大学 | Heat-barrier coating ceramic layer thickness measure new method based on Terahertz Technology |
Non-Patent Citations (3)
| Title |
|---|
| TETSUO IWATA ET AL.: "Prediction of the Thickness of a Thin Paint Film by Applying a Modified Partial-Least-Squares-1 Method to Data Obtained in Terahertz Reflectometry", JOURNAL OF INFRARED, MILLIMETER, AND TERAHERTZ WAVES, pages 646 * |
| 曹丙花;王伟;范孟豹;韦忠亮;: "基于波长选择的纸页厚度太赫兹时域光谱检测新方法", 光谱学与光谱分析, no. 09 * |
| 邢学文;刘松;许德刚;钱凯俊;: "基于偏最小二乘法的高光谱水面油膜厚度估算", 国土资源遥感, no. 02 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114894105A (en) * | 2022-05-16 | 2022-08-12 | 西南科技大学 | Method and system for measuring thickness of non-metallic material in atmospheric environment |
| CN115290597A (en) * | 2022-10-08 | 2022-11-04 | 首都师范大学 | Terahertz technology-based method and system for detecting coating adhesion-free defect |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108519059B (en) | Thermal barrier coating multi-layered thickness detection method based on reflection-type terahertz time-domain spectroscopic technology | |
| US9304046B2 (en) | Sensor system and method for characterizing a coated body | |
| CN114427838A (en) | Method and system for predicting and evaluating thickness of medium based on reflection terahertz spectrum | |
| CN111553877B (en) | Damage identification and service life evaluation method based on terahertz ceramic matrix composite blade | |
| CN111998783B (en) | Reflection type terahertz time-domain spectrum thickness measurement method based on evolutionary optimization algorithm | |
| CN107764195B (en) | Coating thickness detection method based on Fresnel matrix THz wave propagation model | |
| US6184528B1 (en) | Method of spectral nondestructive evaluation | |
| Xu et al. | Terahertz nondestructive quantitative characterization for layer thickness based on sparse representation method | |
| Zhang et al. | THz imaging technique for nondestructive analysis of debonding defects in ceramic matrix composites based on multiple echoes and feature fusion | |
| Luo et al. | Thin thermally grown oxide thickness detection in thermal barrier coatings based on SWT-BP neural network algorithm and terahertz technology | |
| CN105333841A (en) | Metal surface roughness detection device and method based on reflection-type terahertz time-domain spectroscopy | |
| Zhai et al. | Thickness characterization of multi-layer coated steel by terahertz time-of-flight tomography | |
| Tu et al. | Non-destructive evaluation of hidden defects beneath the multilayer organic protective coatings based on terahertz technology | |
| CN111122585B (en) | FDTD-based material microdefect terahertz nondestructive testing simulation method | |
| Balageas | Defense and illustration of time-resolved pulsed thermography for NDE | |
| Kim et al. | Non-destructive detection of thin micro-defects in glass reinforced polymer composites using a terahertz electro–magnetic wave based on a convolution neural network | |
| Cassar et al. | Iterative tree algorithm to evaluate terahertz signal contribution of specific optical paths within multilayered materials | |
| Sun et al. | THzResNet: A physics-inspired two-stream residual network for thermal barrier coating thickness measurement | |
| Zilidou et al. | Suppression of front and back surface reflections in ultrasonic analytic-signal responses from composites | |
| Tu et al. | Quality evaluation of organic protective paints using terahertz pulse imaging technology based on wavelet packet energy method | |
| Jiang et al. | Terahertz Non-destructive testing and imaging of corrosion in coated steel plates | |
| Sun et al. | Physics-based deep learning framework for terahertz thickness measurement of thermal barrier coatings | |
| Picot et al. | Terahertz Paint Thickness Measurements: from lab to automotive and aeronautics industry | |
| Tu et al. | Rapid diagnosis of corrosion beneath epoxy protective coating using non-contact THz-TDS technique | |
| Nguyen et al. | Non-destructive measurement of thickness and refractive index of multilayer coating on metal substrate |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |