Material aging laser multi-dimensional detection method and device based on physical information fusion

Abstract

The invention discloses a material aging laser multi-dimensional detection method and device based on physical information fusion, and the method comprises the steps: constructing a heat-force-sound-light-plasma multi-physical field coupling model through combining multi-dimensional information, such as heat, force, sound, light and plasma, under the action of laser; the interaction between the laser and the material, the interaction between the laser and the plasma and the interaction between the plasma and the atmosphere medium are comprehensively considered, and the physical and chemical characteristic changes of the material can be comprehensively monitored, so that the detection precision and reliability are improved; by introducing three analysis methods of correlation assessment, contribution assessment and importance assessment, the matching degree between each kind of detectable physical information and the aging characteristics is systematically assessed. In the process, the relationship between the aging characteristics and the physical information is quantified through various mathematical models, so that the physical information which has the greatest influence on the aging characteristics is accurately selected, and the accuracy and the reliability of aging state prediction are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of material aging detection, and particularly relates to a laser multi-dimensional detection method and device for material aging based on physical information fusion. Background Art

[0002] With the aging of materials during long-term use, especially in the fields of nuclear energy, aerospace, etc., how to accurately evaluate the aging state of materials and predict their service life has become an important topic in materials science and engineering technology. Traditional material aging detection methods mainly rely on microscopic analysis and mechanical property destructive tests. These methods require offline operation and cause damage to materials, resulting in low detection efficiency and limited data acquisition, making it difficult to achieve accurate material aging assessment and life prediction.

[0003] In recent years, non-destructive testing methods based on laser technology have received extensive attention, especially the application of laser-induced breakdown spectroscopy and laser ultrasonic technology in material aging detection. Laser technology has the advantages of real-time, non-contact, and high spatial resolution, and can effectively overcome the limitations of traditional methods. Laser-induced breakdown spectroscopy technology can provide information on material chemical composition, mechanical structure, and microscopic damage by analyzing the atomic / ionic / molecular emission spectra generated by the interaction between laser and materials. Laser ultrasonic technology uses lasers to excite ultrasonic waves inside materials and evaluates the mechanical properties and internal damage of materials by monitoring the propagation characteristics of ultrasonic waves. During the aging process of materials, their internal physical and chemical characteristics change, which significantly affects the interaction process between laser and aging materials.

[0004] Although laser non-destructive testing technology has significant advantages in material aging monitoring, the current technology still focuses on the detection of single parameters and the pure data fusion level based on machine learning, making it difficult to comprehensively evaluate the aging state of materials and the changes in physical and chemical properties. Summary of the Invention

[0005] The purpose of the present invention is to provide a laser multi-dimensional detection method and device for material aging based on physical information fusion, so as to solve the technical problem that in the prior art, the laser non-destructive testing technology is difficult to comprehensively evaluate the aging state of materials and the changes in physical and chemical properties at the level of single parameter detection and pure data fusion based on machine learning in material aging monitoring.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: A laser multi-dimensional detection method for material aging based on physical information fusion, comprising the following steps: According to the distribution range of aging characteristics, divide the existing sample library into multiple sub-libraries, traverse the multiple sub-libraries in the sample library, use one of the sub-libraries as the validation set, and the remaining sub-libraries as the training set; Determine the discrete / continuous mathematical model corresponding to the aging characteristics in the training set. Based on the discrete / continuous mathematical model, establish a multi-physical field coupling model of heat-mechanics-acoustics-optics-plasma under laser action; Extract various physical information through the output of the multi-physical field coupling model of heat-mechanics-acoustics-optics-plasma, and obtain the physical constraints between the aging characteristics and various physical information; Analyze the matching degree between the aging characteristics and various physical information, and screen out the physical information set from various physical information according to the matching degree from high to low; Introduce the physical constraints between the aging characteristics and various physical information, construct a regression model with hard-coded physical constraints, and obtain the calibration curve between the aging characteristics and the physical information set based on the regression model with hard-coded physical constraints. The calibration curve is used to predict the aging characteristics of the validation set sub-library; Calculate the coincidence degree between the aging sample to be measured and the aging samples corresponding to the validation set in the sample library, and use the calibration curve corresponding to the validation set with the highest coincidence degree to predict the aging characteristics of the aging sample to be measured.

[0007] Furthermore, the aging characteristics include physical characteristics and chemical characteristics. The physical characteristics include mechanical characteristics, thermal characteristics, optical characteristics and mass characteristics. The chemical characteristics include element composition and element content The discrete / continuous mathematical model includes the ideal elastoplastic model, bilinear hardening elastoplastic model, power hardening model or JC constitutive model selected for mechanical characteristics, the isotropic model or anisotropic model selected for thermal characteristics, optical characteristics and mass characteristics, and the stoichiometric equilibrium model or non-stoichiometric equilibrium model selected for chemical characteristics; The physical information includes ablation information, stress information, plasma information, spectral information and acoustic wave information.

[0008] Furthermore, the multi-physical field coupling model of heat-mechanics-acoustics-optics-plasma includes the modeling of the temperature field and stress field inside the aging sample, and the modeling of the flow field and radiation field of the plasma in the atmosphere medium.

[0009] Furthermore, the evaluation of the matching degree between the aging characteristics and various physical information includes correlation evaluation, contribution evaluation and importance evaluation.

[0010] Furthermore, in the correlation evaluation, the evaluation object is the linear / exponential / logarithmic / FFT / power transformation correlation between the aging characteristic and each physical information; In the contribution degree evaluation, if the physical constraint relationship between the aging characteristics and multiple physical information can be analyzed and explicitly calculated, the contribution degree of the physical information to the aging characteristics is the analytical gradient value at the value of the aging characteristics; if the physical constraint relationship between the aging characteristics and multiple physical information cannot be analyzed and explicitly calculated, a polynomial regression model is established between the aging characteristics and the physical information, and the contribution degree of the physical information to the aging characteristics is the numerical gradient value at the value of the aging characteristics on the regression curve; In the importance evaluation, first, a LASSO regression model is established between multiple physical information and aging characteristics. The regularization parameter is initially set to 0. Next, the value of the regularization parameter is continuously increased, and the physical information items with regression coefficients less than 0.1 in the LASSO regression model are removed. Each time a physical information item is removed, the regression is performed again until the regression coefficients of all physical information items are greater than 0.1. At this time, the importance of each physical information item to the aging characteristics is the corresponding regression coefficient; If the matching degrees are the same, the physical information with the largest analytical / numerical gradient value in the contribution degree evaluation is placed in the front.

[0011] Further, the steps for the regression model based on hard-coded physical constraints to obtain the calibration curve between the aging characteristics and the physical information set are as follows: In the regression model of hard-coded physical constraints, each physical information in the physical information set is transformed into an operator with the optimal correlation with the aging characteristics; A calibration curve between the aging characteristics and the physical information set is established through sparse regression.

[0012] Further, the steps for calculating the coincidence degree between the aging sample to be measured and the aging samples corresponding to the validation set in the sample library are as follows: According to the physical information set corresponding to the validation set, obtain the physical information set of the aging sample to be measured of the same type as the validation set, and compare the coincidence degree between the two physical information sets; After obtaining the coincidence degrees between the aging sample to be measured and all sub-libraries in the sample library, select the sub-library with the highest coincidence degree as the calibration curve corresponding to the validation set to predict the aging characteristics of the aging sample to be measured.

[0013] In a second aspect, the present invention provides a laser multi-dimensional detection system for material aging based on physical information fusion, including a division module, a model construction module, a model output module, an analysis module, a relationship acquisition module, and a prediction module, where: The division module: is used to divide the existing sample library into multiple sub-libraries according to the distribution interval of the aging characteristics, traverse the multiple sub-libraries in the sample library, and use one of the sub-libraries as the validation set and the remaining sub-libraries as the training set; Model construction module: used to determine the discrete / continuous mathematical model corresponding to the aging characteristics in the training set, and based on the discrete / continuous mathematical model, establish a multi-physical field coupling model of heat-force-acoustic-optical-plasma under the action of laser; Model output module: used to extract various physical information through the output of the multi-physical field coupling model of heat-force-acoustic-optical-plasma, and obtain the physical constraints between the aging characteristics and various physical information; Analysis module: used to analyze the matching degree between the aging characteristics and various physical information, and screen out the physical information set from various physical information according to the matching degree from high to low; Relationship acquisition module: used to introduce the physical constraints between the aging characteristics and various physical information, construct a regression model with hard-coded physical constraints, and obtain a calibration curve between the aging characteristics and the physical information set based on the regression model with hard-coded physical constraints, and the calibration curve is used to predict the aging characteristics of the validation set sub-library; Prediction module: calculates the coincidence degree between the aging sample to be measured and the aging sample corresponding to the validation set in the sample library, and uses the calibration curve corresponding to the validation set with the highest coincidence degree to predict the aging characteristics of the aging sample to be measured.

[0014] In a third aspect, a terminal device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0015] In a fourth aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0016] Compared with the prior art, the present invention has the following beneficial technical effects: The present invention provides a multi-dimensional laser detection method for material aging based on physical information fusion. By combining multi-dimensional information such as heat, force, sound, light, and plasma under the action of laser, a multi-physical field coupling model of heat-force-acoustic-optical-plasma is constructed, comprehensively considering the interactions between laser and material, laser and plasma, and plasma and atmosphere medium, which can comprehensively monitor the physical and chemical characteristic changes of materials, thereby improving the detection accuracy and reliability; by introducing three analysis methods of correlation evaluation, contribution evaluation, and importance evaluation, the matching degree between each detectable physical information and the aging characteristics is systematically evaluated. This process quantifies the relationship between the aging characteristics and the physical information through various mathematical models, thereby accurately selecting the physical information that has the greatest impact on the aging characteristics, and improving the accuracy and reliability of aging state prediction.

[0017] Preferably, different mathematical models are selected for different aging characteristics, improving the modeling accuracy.

[0018] Preferably, physical features with higher correlation are screened through correlation, contribution degree, and importance evaluation, reducing the computational complexity.

[0019] Preferably, by cross - validating the aging samples to be measured with multiple sub - libraries, the deviation of a single validation set is avoided, enhancing the robustness of the detection results. Description of the Drawings

[0020] Figure 1 It is a flowchart of a laser multi - dimensional detection method for material aging based on physical information fusion in an embodiment of the present invention; Figure 2 It is a flowchart for obtaining a calibration curve in the laser multi - dimensional detection method for material aging based on physical information fusion of the present invention; Figure 3 It is a classification of aging characteristics detectable by the present invention; Figure 4 It is a classification of various physical information options of the present invention; Figure 5 It is a flowchart of the use of a laser multi - dimensional detection device for material aging based on physical information fusion in an embodiment of the present invention; Figure 6 It is a schematic diagram of the stress - strain curve of an aging stainless - steel sample obtained by using a bilinear hardening elastoplastic model to describe its mechanical characteristics; Figure 7 It is a schematic diagram of the elemental composition and the mass fraction of each element in the molten state under laser action of an aging stainless - steel sample obtained by using a non - stoichiometric equilibrium model to describe its chemical characteristics; Figure 8 It is a schematic diagram of the contribution degree ranking between the aging characteristics of an aging stainless - steel sample and 8 physical information obtained by using a polynomial regression model; Figure 9 It is a prediction result graph of the average modulus of an aging stainless - steel sample obtained by applying the present invention; Figure 10 It is a prediction result graph of the Mn element concentration of an aging stainless - steel sample obtained by applying the present invention. Detailed Embodiments

[0021] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0022] In the description of the present invention, it should be understood that the terms "include" and "comprise" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0023] It should also be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0024] It should be further understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. For example, A and / or B can represent three cases: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in the present invention generally represents an "or" relationship between the associated objects before and after.

[0025] It should be understood that although the terms first, second, third, etc. may be used in the embodiments of the present invention to describe preset ranges, etc., these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range can also be referred to as the second preset range, and similarly, the second preset range can also be referred to as the first preset range.

[0026] Depending on the context, the word "if" as used herein can be interpreted as "when" or "while" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detected (stated condition or event)" can be interpreted as "when determined" or "in response to determining" or "when detected (stated condition or event)" or "in response to detecting (stated condition or event)".

[0027] Structural schematic diagrams according to the disclosed embodiments of the present invention are shown in the drawings. These figures are not drawn to scale, where for the purpose of clear expression, some details are enlarged and some details may be omitted. The shapes of the various regions and layers shown in the figures and their relative sizes and positional relationships are only exemplary, and may actually deviate due to manufacturing tolerances or technical limitations. Those skilled in the art can additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0028] The present invention will be further described in detail below with reference to the drawings: As Figure 1 shown, a laser multi-dimensional detection method for material aging based on the fusion of physical information includes the following steps: Step 1: According to the distribution range of aging characteristics, divide the existing sample library into multiple sub-libraries, traverse the multiple sub-libraries in the sample library, use one of the sub-libraries as the validation set, and the remaining sub-libraries as the training set: Specifically, traversing the multiple sub-libraries in the sample library means taking all the sub-libraries in the sample library as the validation set, and the number of validation sets is 1 sub-library each time, and the rest are used as the training set.

[0029]

[0030] Step 2: Determine the discrete / continuous mathematical model corresponding to the aging characteristics in the training set, and based on the discrete / continuous mathematical model, establish a multi-physical field coupling model of heat-mechanics-sound-light-plasma under the action of laser: The aging characteristics include physical characteristics and chemical characteristics. The physical characteristics include mechanical characteristics, thermal characteristics, optical characteristics and mass characteristics. The chemical characteristics include element composition and element content; The discrete / continuous mathematical model includes the ideal elastoplastic model, bilinear hardening elastoplastic model, power hardening model or JC constitutive model selected for mechanical characteristics, the isotropic model or anisotropic model selected for thermal characteristics, optical characteristics and mass characteristics, and the stoichiometric equilibrium model or non-stoichiometric equilibrium model selected for chemical characteristics.

[0031] Step 3: Extract various physical information through the output of the multi-physical field coupling model of heat-mechanics-sound-light-plasma, and obtain the physical constraints between the aging characteristics and various physical information: The multi-physical field coupling model of heat-mechanics-sound-light-plasma under the action of laser involves the interactions between laser and solid samples, laser and plasma generated by samples, and plasma and atmosphere media, including the modeling of temperature field and stress field inside the samples, and the modeling of flow field and radiation field of plasma in atmosphere media; Optionally, the physical information includes ablation information, stress information, plasma information, spectral information and acoustic wave information.

[0032] Step 4: Analyze the matching degree between the aging characteristics and various physical information, and screen out the physical information set from various physical information according to the matching degree from high to low; The evaluation of the matching degree between the aging characteristics and various physical information includes correlation evaluation, contribution evaluation and importance evaluation; In the correlation evaluation, the evaluation object is the linear / exponential / logarithmic / FFT / power transformation correlation between the aging characteristic and each physical information; In the contribution degree evaluation, if the physical constraint relationship between the aging characteristics and multiple physical information can be analyzed and explicitly calculated, the contribution degree of the physical information to the aging characteristics is the analytical gradient value at the value of the aging characteristics; if the physical constraint relationship between the aging characteristics and multiple physical information cannot be analyzed and explicitly calculated, a polynomial regression model between the aging characteristics and the physical information is established, and the contribution degree of the physical information to the aging characteristics is the numerical gradient value at the value of the aging characteristics on the regression curve; In the importance evaluation, first establish a LASSO regression model between multiple physical information and aging characteristics. The regularization parameter is initially set to 0. Next, continuously increase the value of the regularization parameter, and remove the physical information items with regression coefficients less than 0.1 in the LASSO regression model. Re - perform the regression every time an item of physical information is removed until the regression coefficients of all physical information items are greater than 0.1. At this time, the importance of each item of physical information to the aging characteristics is the corresponding regression coefficient; If the matching degrees are the same, the physical information with the largest analytical / numerical gradient value in the contribution degree evaluation is placed in the front.

[0033] Step 5: Introduce the physical constraints between the aging characteristics and multiple physical information, construct a regression model with hard - coded physical constraints, and obtain the calibration curve between the aging characteristics and the physical information set based on the regression model with hard - coded physical constraints. The calibration curve is used to predict the aging characteristics of the validation set sub - library; The steps to obtain the calibration curve between the aging characteristics and the physical information set based on the regression model with hard - coded physical constraints are as follows: In the regression model with hard - coded physical constraints, each type of physical information in the physical information set is transformed into an operator with the optimal correlation with the aging characteristics; Establish the calibration curve between the aging characteristics and the physical information set through sparse regression.

[0034] Step 6: Calculate the coincidence degree between the aging sample to be measured and the aging samples corresponding to the validation set in the sample library, and use the calibration curve corresponding to the validation set with the highest coincidence degree to predict the aging characteristics of the aging sample to be measured.

[0035] The steps to calculate the coincidence degree between the aging sample to be measured and the aging samples corresponding to the validation set in the sample library are as follows: According to the physical information set corresponding to the validation set, obtain the physical information set of the same type as the aging sample to be measured, and compare the coincidence degree between the two physical information sets; After obtaining the coincidence degrees between the aging sample to be measured and all sub - libraries in the sample library, use the sub - library with the highest coincidence degree as the calibration curve corresponding to the validation set to predict the aging characteristics of the aging sample to be measured.

[0036] Specifically, the method for obtaining the physical information set of the aging sample to be measured is as follows: Use a microscopic imaging system, a thermal imaging system, etc. to obtain ablation information, such as an optical microscope, a thermocouple array. Use an ultrasonic detection system, a displacement sensing system, etc. to obtain stress information, such as a laser interferometer, an air-coupled ultrasonic system. Use a plasma fast imaging system, a laser shadow imaging system, etc. to obtain plasma information, such as ICCD fast imaging, a streak camera. Use a spectrometer to obtain spectral information, such as an echelle grating spectrometer, a Raman spectrometer. Use an acoustic emission sensor, a pressure sensor to measure acoustic wave information, such as a shock wave high-frequency pressure sensor, a needle hydrophone.

[0037] As Figure 2 shown, in this embodiment, a laser multi-dimensional detection method for material aging based on physical information fusion is provided, including the following steps: 1) According to the distribution interval of known aging characteristics, divide the existing sample library into K sub-libraries of equal size, and let i = 1; Specifically, the total number of samples in the sample library is M , the number of sub-libraries is K , and the number of physical information that can be detected n ∈[1, 0.9×( K - 1)× M / K ; 2) Input the i th sub-library as the validation set, and the remaining K - 1 sub-libraries as the training set; 3) For the distribution interval and physical and chemical properties of the aging characteristics in the training set, select a suitable discrete / continuous mathematical model for this aging characteristic; As Figure 3 shown, the aging characteristics include two categories: physical characteristics and chemical characteristics. Physical characteristics include mechanical characteristics, thermal characteristics, optical characteristics, and mass characteristics; chemical characteristics include element composition and element content; For mechanical characteristics, select various elastic moduli (Young's modulus, shear modulus, bulk modulus, Poisson's ratio, etc.), yield strength, tensile strength, hardness, elongation, reduction of area, and Charpy impact energy; for thermal characteristics, select specific heat capacity, thermal conductivity, melting enthalpy, evaporation enthalpy, melting point, boiling point, and coefficient of thermal expansion; for optical characteristics, select optical absorption coefficient and surface reflectivity; for mass characteristics, select total mass density (solid phase), total mass density (liquid phase), and average atomic mass; for element composition, select the types of elements contained in the sample; for element content, select the mass fraction and mole fraction of each element; The discrete / continuous mathematical models of aging characteristics are as follows: For mechanical characteristics, an ideal elastoplastic model, a bilinear hardening elastoplastic model, a power hardening model, or a JC (Johnson-Cook) constitutive model is selected. For thermal, optical, and mass characteristics, an isotropic model or an anisotropic model is selected. For chemical characteristics, a stoichiometric equilibrium model or a non-stoichiometric equilibrium model is selected.

[0038] 4) Taking this aging characteristic as an input variable, establish a multi-physical field coupling model of heat-mechanics-acoustics-optics-plasma under laser action; The multi-physical field coupling model of heat-mechanics-acoustics-optics-plasma under laser action involves the interactions between the laser and the solid sample, the laser and the plasma generated by the sample, and the plasma and the atmosphere medium. The multi-physical field coupling model includes the modeling of the temperature field and stress field inside the sample, and the modeling of the flow field and radiation field of the plasma in the atmosphere medium.

[0039] 5) Among the output variables of the above multi-physical field coupling model, initially extract m types of detectable physical information, and obtain the physical constraints between this aging characteristic and m types of detectable physical information; As Figure 4 shown, among the output variables of the multi-physical field coupling model, the detectable physical information includes ablation information, stress information, plasma information, spectral information, and acoustic wave information; ablation information includes ablation mass, ablation depth, ablation volume, temperature distribution, and temperature gradient distribution; stress information includes displacement distribution, stress distribution, and strain distribution. Plasma information can be the plasma expansion velocity, expansion position, size, morphology, dimensions, energy, temperature, electron number density, average ionization degree, and number density of ions with different valence states; spectral information includes bremsstrahlung intensity, atomic emission line spectrum radiation intensity, ionic emission line spectrum radiation intensity, and molecular emission band intensity. Acoustic wave information can be the shock wave front pressure, position, and velocity.

[0040] 6) Analyze the matching degree between this aging characteristic and m types of detectable physical information respectively, including three aspects of evaluation: correlation evaluation, contribution degree evaluation, and importance evaluation; In the correlation evaluation, the evaluation object is the correlation of linear / exponential / logarithmic / FFT / power transformation between this aging characteristic and each type of detectable physical information, and the evaluation indicators include the determination coefficient R 2 , root mean square error RMSE, and mean absolute percentage error MAPE; In the contribution degree evaluation, if the physical constraint relationship between the aging feature and the detectable physical information can be analyzed and explicitly calculated, the contribution degree of the detectable physical information to the aging feature is the analytical gradient value at the value of the aging feature; if the physical constraint relationship between the aging feature and the detectable physical information cannot be analyzed and explicitly calculated, a polynomial regression model is established between the aging feature and the detectable physical information, and the contribution degree of the detectable physical information to the aging feature is the numerical gradient value at the value of the aging feature on the regression curve. In the importance evaluation, first establish m a LASSO regression model between the detectable physical information and the aging feature. The regularization parameter is initially set to 0. Next, continuously increase the value of the regularization parameter, remove the items of detectable physical information with regression coefficients less than 0.1 in the LASSO regression model, and re - perform the regression every time an item of detectable physical information is removed until the regression coefficients of all items of detectable physical information are greater than 0.1. At this time, the importance of each item of detectable physical information to the aging feature is the corresponding regression coefficient. 7) According to the evaluation results, screen out the top n types of detectable physical information in descending order of the matching degree; The ranking of the matching degree between each item of detectable physical information and the aging feature is obtained by averaging the rankings of three evaluations: correlation evaluation, contribution degree evaluation, and importance evaluation. If the rankings are the same, the detectable physical information with the largest analytical / numerical gradient value in the contribution degree evaluation is ranked first.

[0041] 8) Based on the regression model with hard - coded physical constraints, establish a calibration curve between the aging feature and n the types of detectable physical information; L i ; In the regression model with hard - coded physical constraints, n each type of detectable physical information is transformed into the operator with the best correlation with the aging feature, and a calibration curve between the aging feature and n the types of detectable physical information is established through sparse regression. L i 。

[0042] 9) Judge whether i ≤ K holds. If it holds, let i = i +1, and repeat the work from step 2) to step 10); if it does not hold, stop the above steps; 10) For each aging feature of the aging sample to be measured, calculate the coincidence degree between the aging samples in it and the aging sample to be measured when each sub - library is used as the validation set separately under this aging feature. When calculating the coincidence degree between the aging sample to be measured and the i th sub - library, let in the i th sub - library, then The physical information that can be detected is X T = { X 1 , X 2 , X 3 , …, X n}, then the aging sample to be tested needs to obtain the corresponding n types of physical information that can be detected, that is X Pi = { X 1i , X 2i , X 3i , …, X ni}. Define that under this aging characteristic, the coincidence degree between the aging sample to be tested and the i th sub-library is || X T - X Pi ||2 -1 =[( X 1 - X 1i ) 2 +( X 2 - X 2i ) 2 +( X 3 - X 3i ) 2 +…+( X n - X ni ) 2 -1 / 2 ; After obtaining the coincidence degrees between the aging sample to be tested and all sub-libraries, the sub-library with the highest coincidence degree is taken as the calibration curve L established when using it as the validation set i is the calibration curve required for this aging characteristic to be tested. Based on this calibration curve and the corresponding n types of physical information that can be detected, the predicted value of this aging characteristic to be tested is calculated.

[0043] As Figure 5 shown, each aging sample corresponds to multiple aging characteristics. Therefore, different sample libraries are required for predicting different aging characteristics of each aging sample to be tested. Therefore, the prediction steps for all aging characteristics of the aging sample to be tested are as Figure 5 shown.

[0044] ​In an alternative embodiment, as Figure 6 shown, the average modulus of the stainless steel aging samples used is brought into the thermo-mechanical-acoustic-optical-plasma multi-physical field coupling model using a bilinear hardening elastoplastic model. Among the detectable physical information output by the multi-physical field coupling model, ablation information, stress information, plasma information, and spectral information are used to comprehensively evaluate their contribution degrees to the average modulus. In the contribution degree evaluation, their ranking is as Figure 8 shown. In ascending order of the characteristic numbers, they are stress information, electron number density, spectral intensity ratio of Fe II / Fe I, plasma energy, plasma temperature, ablation depth, strain information, and spectral intensity ratio of Cr I / Fe I. According to the comprehensive matching degree ranking, the prediction effect of the average modulus of all samples obtained by selecting the first 4 detectable physical information is as Figure 9 shown, which can reach 0.9851, proving that the predicted value is very close to the actual value.

[0045] As Figure 7 shown, in this embodiment, the element content of the stainless steel aging samples used is brought into the thermo-mechanical-acoustic-optical-plasma multi-physical field coupling model using a non-stoichiometric equilibrium model. According to the comprehensive matching degree ranking, the prediction effect of the Mn element concentration of all samples obtained by selecting the first 2 detectable physical information is as Figure 10 shown, which can reach 0.9969, also confirming the effectiveness of the method of the present invention.

[0046] In another embodiment of the present invention, there is also provided a laser multi-dimensional detection system for material aging based on physical information fusion, including a division module, a model construction module, a model output module, an analysis module, a relationship acquisition module, and a prediction module, wherein: The division module: is used to divide the existing sample library into multiple sub-libraries according to the distribution range of the aging characteristics, traverse the multiple sub-libraries in the sample library, use one of the sub-libraries as the verification set, and the remaining sub-libraries as the training set; The model construction module: is used to determine the discrete / continuous mathematical model corresponding to the aging characteristics in the training set, and based on the discrete / continuous mathematical model, establish a thermo-mechanical-acoustic-optical-plasma multi-physical field coupling model under laser action; The model output module: is used to extract various physical information through the output of the thermo-mechanical-acoustic-optical-plasma multi-physical field coupling model, and obtain the physical constraints between the aging characteristics and various physical information; The analysis module: is used to analyze the matching degree between the aging characteristics and various physical information, and screen out the physical information set from various physical information according to the matching degree from high to low; Relationship acquisition module: It is used to introduce the physical constraints between aging characteristics and various physical information, construct a regression model with hard-coded physical constraints, and obtain a calibration curve between the aging characteristics and the set of physical information based on the regression model with hard-coded physical constraints. The calibration curve is used to predict the aging characteristics of the validation set sub-library. Prediction module: Calculate the coincidence degree between the aging sample to be measured and the aging samples corresponding to the validation set in the sample library, and use the calibration curve corresponding to the validation set with the highest coincidence degree to predict the aging characteristics of the aging sample to be measured.

[0047] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0048] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be realized by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0049] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0050] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for realizing the functions in Figure 1 one process or multiple processes and / or blocks Figure 1Steps of the functions specified in one or more boxes.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the scope of its protection. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: after reading the present invention, those skilled in the art can still make various changes, modifications or equivalent replacements to the specific implementation manners of the invention, but these changes, modifications or equivalent replacements are all within the scope of the claims of the invention pending approval.

Claims

1. A laser multi-dimensional detection method for material aging based on the fusion of physical information, characterized in that, It includes the following steps: According to the distribution interval of the aging characteristics, divide the existing sample library into multiple sub-libraries, traverse the multiple sub-libraries in the sample library, take one of the sub-libraries as the validation set, and the remaining sub-libraries as the training set; Determine the discrete / continuous mathematical model corresponding to the aging characteristics in the training set. Based on the discrete / continuous mathematical model, establish a multi-physical field coupling model of heat-mechanics-sound-light-plasma under laser action; Extract various physical information through the output of the multi-physical field coupling model of heat-mechanics-sound-light-plasma, and obtain the physical constraints between the aging characteristics and various physical information; Analyze the matching degree between the aging characteristics and various physical information, and screen out the physical information set from various physical information according to the matching degree from high to low; Introduce the physical constraints between the aging characteristics and various physical information, construct a regression model with hard-coded physical constraints, and obtain the calibration curve between the aging characteristics and the physical information set based on the regression model with hard-coded physical constraints. The calibration curve is used to predict the aging characteristics of the validation set sub-library; Calculate the coincidence degree between the aging sample to be measured and the aging sample corresponding to the validation set in the sample library, and use the calibration curve corresponding to the validation set with the highest coincidence degree to predict the aging characteristics of the aging sample to be measured.

2. The method for multi-dimensional laser detection of material aging based on physical information fusion according to claim 1, wherein The aging characteristics include physical characteristics and chemical characteristics, The physical characteristics include mechanical characteristics, thermal characteristics, optical characteristics and mass characteristics, and the chemical characteristics include element composition and element content; The discrete / continuous mathematical model includes the ideal elastoplastic model, bilinear hardening elastoplastic model, power hardening model or JC constitutive model selected for mechanical characteristics, the isotropic model or anisotropic model selected for thermal characteristics, optical characteristics and mass characteristics, and the stoichiometric equilibrium model or non-stoichiometric equilibrium model selected for chemical characteristics; The physical information includes ablation information, stress information, plasma information, spectral information and acoustic wave information.

3. A laser multi-dimensional detection method for material aging based on physical information fusion according to claim 1, characterized in that, The multi-physical field coupling model of heat-mechanics-sound-light-plasma includes the modeling of the temperature field and stress field inside the aging sample, and the modeling of the flow field and radiation field of the plasma in the atmosphere medium.

4. A laser multi-dimensional detection method for material aging based on physical information fusion according to claim 1, characterized in that The evaluation of the matching degree between the aging characteristics and various physical information includes correlation evaluation, contribution evaluation and importance evaluation.

5. A laser multi-dimensional detection method for material aging based on physical information fusion according to claim 4, characterized in that, In the correlation evaluation, the evaluation object is the linear / exponential / logarithmic / FFT / power transformation correlation between the aging characteristic and each physical information; In the contribution evaluation, if the physical constraint relationship between the aging characteristic and various physical information can be analyzed and explicitly calculated, the contribution of the physical information to the aging characteristic is the analytical gradient value at the value of the aging characteristic; if the physical constraint relationship between the aging characteristic and various physical information cannot be analyzed and explicitly calculated, establish a polynomial regression model between the aging characteristic and the physical information, and the contribution of the physical information to the aging characteristic is the numerical gradient value at the value of the aging characteristic on the regression curve; In the importance evaluation, first, a LASSO regression model between various physical information and aging characteristics is established. The regularization parameter is initially set to 0. Next, the value of the regularization parameter is continuously increased, and the physical information items with regression coefficients less than 0.1 in the LASSO regression model are removed. After removing each physical information item, re - regression is performed until the regression coefficients of all physical information items are greater than 0.

1. At this time, the importance of each physical information item to the aging characteristic is the corresponding regression coefficient; If the matching degrees are the same, the physical information with the largest analytical / numerical gradient value in the contribution degree evaluation is placed in the front.

6. The laser multi-dimensional detection method for material aging based on physical information fusion according to claim 1, wherein The steps for the regression model based on hard - coded physical constraints to obtain the calibration curve between the aging characteristic and the physical information set are as follows: In the regression model with hard - coded physical constraints, each physical information in the physical information set is transformed into an operator with the optimal correlation with the aging characteristic; A calibration curve between the aging characteristic and the physical information set is established through sparse regression.

7. A laser multi-dimensional detection method for material aging based on physical information fusion according to claim 1, characterized in that The steps for calculating the coincidence degree between the aging sample to be measured and the aging samples corresponding to the validation set in the sample library are as follows: According to the physical information set corresponding to the validation set, obtain the physical information set of the aging sample to be measured of the same type as the validation set, and compare the coincidence degree between the two physical information sets; After obtaining the coincidence degrees between the aging sample to be measured and all sub - libraries in the sample library, select the sub - library with the highest coincidence degree as the calibration curve corresponding to the validation set to predict the aging characteristics of the aging sample to be measured.

8. A laser multi-dimensional detection system for material aging based on the fusion of physical information, characterized in that, It includes a division module, a model construction module, a model output module, an analysis module, a relationship acquisition module, and a prediction module, where: The division module: is used to divide the existing sample library into multiple sub - libraries according to the distribution interval of the aging characteristic, traverse the multiple sub - libraries in the sample library, take one of the sub - libraries as the validation set, and the remaining sub - libraries as the training set; The model construction module: is used to determine the discrete / continuous mathematical model corresponding to the aging characteristic in the training set, and based on the discrete / continuous mathematical model, establish a multi - physical - field coupling model of heat - force - sound - light - plasma under laser action; The model output module: is used to extract various physical information through the output of the multi - physical - field coupling model of heat - force - sound - light - plasma, and obtain the physical constraints between the aging characteristic and various physical information; The analysis module: is used to analyze the matching degree between the aging characteristic and various physical information, and screen out the physical information set from various physical information according to the matching degree from high to low; The relationship acquisition module: is used to introduce the physical constraints between the aging characteristic and various physical information, construct a regression model with hard - coded physical constraints, obtain the calibration curve between the aging characteristic and the physical information set based on the regression model with hard - coded physical constraints, and the calibration curve is used to predict the aging characteristics of the validation set sub - library; The prediction module: calculates the coincidence degree between the aging sample to be measured and the aging samples corresponding to the validation set in the sample library, and takes the calibration curve corresponding to the validation set with the highest coincidence degree to predict the aging characteristics of the aging sample to be measured.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 - 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

Images