Method, system and equipment for rapidly detecting bending strength of material and medium

Abstract

The invention discloses a method, a system, equipment and a medium for rapidly detecting the bending strength of a material, and belongs to the technical field of bending strength detection.The method comprises the steps that spectral data of a sample to be detected is obtained, the spectral data is preprocessed, and preprocessed spectral data is obtained; performing spectral line identification on the preprocessed spectral data, determining an ion spectral line and an atomic spectral line of a target element, and determining a spectral characteristic quantity of the target element according to the ion spectral line and the atomic spectral line; obtaining a standard spectrum characteristic quantity based on a standard sample, and obtaining a calibration model according to the standard spectrum characteristic quantity; and determining the bending strength value of the to-be-detected sample according to the spectral characteristic quantity and the calibration model. According to the invention, the spectral data of the to-be-detected sample is obtained by using the laser-induced breakdown spectroscopy technology, so that irreversible damage to the original structure is avoided; the ion atomic spectral line intensity ratio is selected as the spectral characteristic quantity, and the ionization characteristic of the material in the laser ablation process can be reflected.

Description

Technical Field

[0001] This invention relates to the field of bending strength testing technology, and in particular to a method, system, equipment and medium for rapidly testing the bending strength of materials. Background Technology

[0002] Bending strength plays a crucial role in engineering design, structural safety, material selection, and quality control, especially in the mechanical, construction, and transportation sectors, where it is an indispensable core performance indicator. In power facilities, both metallic and non-metallic materials are widely used. These materials are constantly exposed to the natural environment and electric fields, making them highly susceptible to corrosion and aging, leading to gradual performance degradation. This is particularly true for the numerous structural components in transmission lines, which face even more stringent requirements regarding their mechanical properties. For example, high-temperature, high-humidity, and high-salinity climatic conditions significantly accelerate the material degradation process. Simultaneously, the strong wind loads and environmental impacts brought by extreme weather events such as typhoons further test the bending strength of power facility materials. With the continuous expansion of the power grid and the advancement of smart grid construction, traditional operation and maintenance models are no longer sufficient to meet the needs of safe and reliable operation of power facilities. Bending strength testing not only provides a scientific basis for equipment condition assessment and hazard identification but also provides important data support for material selection, life prediction, and maintenance optimization. It is of great significance for achieving full life-cycle management of power facilities, reducing operation and maintenance costs, and improving the safety and stability of the power system.

[0003] Currently, conventional bending strength testing methods require cutting the sample into strips of specified dimensions and conducting destructive testing using an electronic universal testing machine. These experiments not only require sampling and cutting of actual operating equipment, leading to damage to existing components, but also suffer from complex operation, long cycles, and high costs, making them unsuitable for the practical needs of rapid, large-scale on-site testing. Therefore, traditional testing methods are generally only applicable during the selection and development of new materials and cannot be used for routine condition monitoring of in-service or critical equipment. In contrast, laser-induced breakdown spectroscopy (LAS) technology can perform non-contact, quasi-non-destructive rapid testing of materials, enabling the identification of the composition and condition of materials in equipment such as enclosures. This reduces manpower and material consumption and improves testing efficiency and accuracy, providing an effective means for health monitoring and fault early warning of in-service equipment. It is expected to overcome the limitations of traditional testing and better serve the safe operation and intelligent management of power systems. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the present application solves the technical problem of how to solve the problem that the conventional bending strength detection method is time-consuming and laborious and cannot realize rapid detection, by using the laser-induced breakdown spectroscopy (LIBS) technology, ablation of the material surface by high-energy laser pulses, generation of high-temperature plasma, analysis of the radiation spectrum generated by the transition of the particles in the plasma cooling, and establishment of a method for rapidly detecting the bending strength of the material.

[0006] To solve the above technical problems, the present application provides the following technical solutions: a method for rapidly detecting the bending strength of a material, comprising the following steps,

[0007] Obtaining spectral data of a sample to be measured, pre-processing the spectral data to obtain pre-processed spectral data, identifying spectral lines of the pre-processed spectral data to determine ion spectral lines and atomic spectral lines of a target element, and determining spectral feature quantities of the target element according to the ion spectral lines and the atomic spectral lines; obtaining standard spectral feature quantities based on standard samples, and obtaining a calibration model according to the standard spectral feature quantities; and determining the bending strength value of the sample to be measured according to the spectral feature quantities and the calibration model.

[0008] As a preferred scheme of the method for rapidly detecting the bending strength of a material, the method comprises the following steps: determining the spectral peak position according to the pre-processed spectral data; comparing the spectral peak position with an atomic spectral database to determine the element attribution of the spectral data; and determining the ion spectral lines and the atomic spectral lines of the target element according to the element attribution of the spectral data. The beneficial effects of the preferred scheme are that the ion spectral lines and the atomic spectral lines of the target element are attributed by comparing the spectral peak position with the atomic spectral database, which can eliminate the subjective error of manual identification, improve the accuracy and reliability of the spectral line attribution, and ensure the accuracy of the spectral feature quantity calculation.

[0009] As a preferred scheme of the method for rapidly detecting the bending strength of a material, the method comprises the following steps: obtaining the spectral line intensity of the ion spectral lines and the spectral line intensity of the atomic spectral lines according to the ion spectral lines and the atomic spectral lines of the target element; and obtaining the spectral feature quantities of the target element based on the spectral line intensity of the ion spectral lines and the spectral line intensity of the atomic spectral lines.

[0010] As a preferred scheme of the method for rapidly detecting the bending strength of a material, in the method, the standard sample-based standard spectral feature quantity is obtained, and a calibration model is obtained according to the standard spectral feature quantity, including: obtaining spectral data of standard samples with known bending strength values; determining a standard spectral feature quantity according to the spectral data of the standard samples; and performing linear regression analysis on the standard spectral feature quantity and the known bending strength values to obtain a calibration model. The preferred scheme has the beneficial effect that the calibration model is established by obtaining the standard samples with known bending strength values and performing linear regression analysis, so that a quantitative relationship between the spectral feature quantity and the bending strength is established, accurate conversion from spectral information to mechanical properties is realized, and the reliability of the detection result is improved.

[0011] As a preferred scheme of the method for rapidly detecting the bending strength of a material, in the method, the spectral data of the sample to be detected are obtained, including: performing laser ablation on the sample to be detected to generate plasma; and collecting emission spectrum of the plasma to obtain spectral data of the sample to be detected.

[0012] As a preferred scheme of the method for rapidly detecting the bending strength of a material, in the method, the spectral data are preprocessed to obtain preprocessed spectral data, including: removing background spectrum in the spectral data; performing baseline correction and noise reduction processing on the spectral data after the background spectrum is removed; and performing normalization processing on the spectral data after the baseline correction and noise reduction processing to obtain preprocessed spectral data.

[0013] As a preferred scheme of the method for rapidly detecting the bending strength of a material, in the method, the spectral feature quantity of the target element is obtained based on the spectral line intensity of the ion spectral line and the spectral line intensity of the atomic spectral line, that is, the ratio of the spectral line intensity of the ion spectral line to the spectral line intensity of the atomic spectral line is calculated to obtain the spectral feature quantity of the target element. The preferred scheme has the beneficial effect that the ratio of the ion spectral line intensity to the atomic spectral line intensity is calculated as the spectral feature quantity, which can effectively reflect the ionization degree characteristics of the material, and the ionization degree is closely related to the microstructure and chemical state of the material, so that accurate characterization and prediction of the bending strength are realized.

[0014] Another object of the present application is to provide a system for rapidly detecting the bending strength of a material.

[0015] To solve the above technical problems, the present application provides the following technical solutions: a system for rapidly detecting the bending strength of a material, comprising: a spectral data module configured to acquire spectral data of a sample to be tested, pre-process the spectral data, and obtain pre-processed spectral data; a spectral feature quantity module configured to identify spectral lines from the pre-processed spectral data, determine ion spectral lines and atomic spectral lines of a target element, and determine spectral feature quantities based on the ion spectral lines and the atomic spectral lines; a calibration model module configured to acquire standard spectral feature quantities based on standard samples, and obtain a calibration model based on the standard spectral feature quantities; and a bending strength module configured to determine a bending strength value of the sample to be tested based on the spectral feature quantities and the calibration model.

[0016] The present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the method for rapidly detecting the bending strength of a material when executing the computer program.

[0017] The present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the method for rapidly detecting the bending strength of a material when executed by a processor.

[0018] The present application has the following advantages: the present application acquires spectral data of a sample to be tested using laser-induced breakdown spectroscopy technology and pre-processes the spectral data, thereby solving the problem of destructive sampling in traditional bending strength detection, using laser ablation to generate plasma to form a microscopic ablation pit on the surface of the material, thereby avoiding irreversible damage to the original structure and achieving quasi-non-destructive detection; automatically identifying spectral lines through a continuous wavelet transform peak-finding algorithm and comparing with an atomic spectral database, selecting an ion-atomic spectral line intensity ratio as a spectral feature quantity, which can reflect the ionization characteristics of the material during laser ablation, is closely related to the microstructure and chemical bonding state of the material, and establishes a scientific internal relationship with the macro bending strength performance, thereby having stronger stability and anti-interference ability; establishing a plurality of known bending strength standard sample libraries, using multi-point average measurement and linear regression analysis to establish a quantitative relationship model between the spectral feature quantity and the bending strength, using an experimental data-driven modeling method to effectively reduce the influence of spectral volatility on modeling accuracy, and using least squares linear regression analysis to ensure the statistical optimality of model parameters, thereby achieving the effect of transitioning from qualitative analysis to quantitative detection; achieving continuous monitoring of a running device, thereby being able to capture real-time changes in the performance of the material and providing an effective technical means for the safe operation and intelligent management of a power system, thereby achieving real-time, continuous, and efficient detection effects that cannot be achieved by traditional detection methods. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings. Among them:

[0020] Figure 1 The overall flowchart of a method for rapidly detecting the bending strength of a material according to an embodiment of the present application.

[0021] Figure 2 The structural schematic diagram of a laser-induced breakdown spectroscopy device according to an embodiment of the present application.

[0022] Figure 3 The structural schematic diagram of a computer device according to an embodiment of the present application.

[0023] Figure 4 The original spectrum schematic diagram according to an embodiment of the present application.

[0024] Figure 5 The spectrum schematic diagram after baseline correction and noise reduction according to an embodiment of the present application.

[0025] Figure 6 The calibration curve diagram of the ion atomic spectral line intensity ratio and the bending strength of a glass fiber reinforced unsaturated polyester molding compound sample according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the above-mentioned objects, features and advantages of the present application more apparent and comprehensible, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0027] In the following description, a lot of specific details are set forth in order to facilitate a full understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, so the present application is not limited to the specific embodiments disclosed below.

[0028] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. The "in one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.

[0029] Embodiment 1, refer to Figure 1 The first embodiment of the present application provides a method for rapidly detecting the bending strength of a material, which comprises the following steps:

[0030] S100: Obtain spectral data of a sample to be tested, and pre-process the spectral data to obtain pre-processed spectral data.

[0031] S200: Perform spectral line identification on the pre-processed spectral data, determine the ion spectral line and the atomic spectral line of the target element, and determine the spectral feature quantity of the target element according to the ion spectral line and the atomic spectral line.

[0032] S300: Obtain standard spectral feature quantities based on standard samples, and obtain a calibration model according to the standard spectral feature quantities.

[0033] S400: Determine the bending strength value of the sample to be tested according to the spectral feature quantity and the calibration model.

[0034] It should be noted that the existing material bending strength detection technology mainly adopts a destructive test method, which needs to cut standard samples from the tested component, and obtains the bending strength data through three-point bending or four-point bending test. Such detection method has obvious limitations: first, it causes irreversible damage to the original structure, especially for key load-bearing components, which may affect the safety of the structure itself; second, the sample preparation process is complex, which requires precise size control and surface treatment, and the detection period is as long as several hours or even several days; third, the cost of single detection is high, including equipment usage fee, labor cost and material loss fee; and the key components in the power system, such as transmission lines, transformer equipment supports and insulators, are subjected to mechanical load, electrical stress and environmental erosion for a long time, and the material performance shows changing characteristics, especially in harsh environments such as coastal areas, high altitudes and industrial pollution, the material degradation speed is accelerated, and the bending strength decay law is complex; due to the destructive characteristics of the traditional detection method, it cannot realize continuous monitoring of the running equipment, and can only be used for offline sampling inspection, which is difficult to capture the real-time change trend of the material performance, and the laser-induced breakdown spectroscopy technology can analyze the material surface plasma emission characteristics to obtain the element composition and microstructure information closely related to the mechanical properties.

[0035] Therefore, in view of the low detection efficiency and insufficient real-time performance, the steps of S100-S400 are used to establish a material bending strength rapid detection method based on laser-induced breakdown spectroscopy characteristics, the ratio of the ion spectral line intensity and the atomic spectral line intensity of the target element is extracted as the spectral feature quantity, a quantitative relationship model between the spectral feature quantity and the bending strength is constructed, and the non-destructive rapid determination of the material bending strength is realized; at the same time, based on the high efficiency and portability of the spectral detection, batch screening and performance degradation warning of the equipment are realized.

[0036] Embodiment 2, refer to Figure 1 and Figure 2For the second embodiment of the present application, based on the above-mentioned embodiment, a method for rapidly detecting the bending strength of a material is provided.

[0037] In the embodiment of the present application, the spectral data of the sample to be measured is acquired in step S100, including the following steps A1-A2:

[0038] A1: Laser ablation is performed on the sample to be measured to generate plasma.

[0039] A2: Collect the emission spectrum of the plasma to obtain the spectral data of the sample to be measured.

[0040] Specifically, in step A1, laser ablation is performed on the sample to be measured to generate plasma, which means that laser-induced breakdown spectroscopy technology is used to perform laser ablation on the sample to be measured to generate plasma on the surface of the sample to be measured.

[0041] Specifically, in step A2, the emission spectrum of the plasma is collected to obtain the spectral data of the sample to be measured, which means that the emission spectrum of the plasma is collected using an optical fiber, wherein the emission spectrum contains the element composition and physical and chemical state information of the ablated sample, providing basic data for subsequent bending strength detection and analysis.

[0042] In an alternative embodiment, the spectral data of the sample to be measured in step S100 can also be obtained by combining Raman spectroscopy technology with surface enhancement technology, using the interaction between laser and material molecular vibration to obtain the molecular structure information of the material, reflecting the internal stress state and micro-defect distribution of the material by analyzing the characteristic peak position and intensity change, thereby establishing a correlation between the bending strength, suitable for non-destructive detection of polymer-based composite materials.

[0043] In another alternative embodiment, the spectral data of the sample to be measured in step S100 can also be obtained by near-infrared spectroscopy technology, using the absorption characteristics of near-infrared light by the material to obtain molecular bonding information, combining chemometrics methods to analyze material component changes and aging degree, establishing a quantitative model of spectral absorption peaks and bending strength, suitable for on-site rapid detection of insulating materials and composite materials in power equipment.

[0044] It should be noted that compared with the traditional detection method, the laser-induced breakdown spectroscopy technology can realize non-contact, quasi-non-destructive and rapid detection of materials, realize the composition and state identification of the box and other equipment materials, reduce the consumption of manpower and material resources, and improve the detection efficiency and accuracy, which provides an effective means for health monitoring and fault warning of in-service equipment, and is expected to break through the limitations of traditional detection and better serve the safe operation and intelligent management of the power system. The laser-induced breakdown spectroscopy technology is to ablate the sample by high-energy pulsed laser, generate plasma on the surface of the sample, and collect the emission spectrum of the plasma by an optical fiber. The emission spectrum contains information such as the elemental composition and physical and chemical state of the ablated sample. The bending strength of the material is mainly determined by the elemental composition, microstructure and other factors, which will directly affect the laser ablation process, the generation of plasma and the spectral characteristics. For example, different component ratios, structural density or internal defects will cause certain spectral characteristics (such as ion / atomic spectral line intensity, spectral line intensity, plasma temperature, electron density, etc.) in the laser-induced breakdown spectroscopy. Therefore, when the experimental parameters of the laser-induced breakdown spectroscopy remain unchanged, laser ablation of samples with different bending strengths can obtain spectral information with differences, and there is often a good linear relationship between certain spectral characteristic quantities and the bending strength, thereby establishing a method for rapidly detecting the bending strength of the material.

[0045] Further, in order to realize laser ablation, a laser-induced breakdown spectroscopy device needs to be built, as shown in Figure 2 The laser-induced breakdown spectroscopy system mainly consists of four parts, including a laser, an optical system, a controller and a spectrometer. By selecting appropriate laser energy, light collection angle and spectrometer delay time, spectral signals with high signal-to-noise ratio and signal-to-background ratio can be obtained.

[0046] In the embodiment of the present application, the spectral data is preprocessed in step S100 to obtain preprocessed spectral data, including the following steps B1-B3:

[0047] B1: Remove the background spectrum in the spectral data.

[0048] B2: Perform baseline correction and noise reduction processing on the spectral data after removing the background spectrum.

[0049] B3: Perform normalization processing on the spectral data after baseline correction and noise reduction processing to obtain preprocessed spectral data.

[0050] Specifically, the background spectrum in the spectral data is removed in step B1, and the specific operation can be to collect the background spectrum at the same time during testing, and subtract the background spectrum from the original spectrum to eliminate the interference of environmental light and instrument noise.

[0051] Specifically, the baseline correction and noise reduction processing of the spectrum data after removing the background spectrum in step B2 can be based on discrete wavelet transform, so that most of the noise is eliminated and the spectral lines from different channels are well corrected.

[0052] Specifically, the normalization processing in step B3 refers to full spectrum area normalization, that is, the spectral line intensity is divided by the total spectral area, so as to eliminate the influence of laser power fluctuation and sample surface state difference on the spectral intensity.

[0053] In an alternative embodiment, the pre-processing of the spectrum data in step S100 can also be performed by a hybrid dimension reduction method combining principal component analysis and independent component analysis, that is, the main variation information of the spectrum data is extracted by principal component analysis, and then the signal components of different sources are separated by independent component analysis, so as to effectively remove instrument drift and environmental interference, while retaining effective information related to material performance, and improving the accuracy of subsequent spectral line identification.

[0054] In another alternative embodiment, the pre-processing of the spectrum data in step S100 can also be performed by a filtering technology combined with empirical mode decomposition method, that is, the filtering parameters are automatically adjusted according to the real-time characteristics of the spectrum signal, the spectrum signal is decomposed into a plurality of intrinsic mode functions, and the noise and interference of different frequency components are processed respectively, so as to realize more accurate signal purification and baseline correction effect.

[0055] It should be noted that the present application obtains the spectrum data of the sample to be measured by laser-induced breakdown spectroscopy technology and performs pre-processing, so as to realize rapid, non-destructive detection of material composition and physical and chemical state information; compared with the traditional destructive bending strength detection method, the present application solves the problem of damage to the original component caused by sampling and cutting of the actual running equipment, especially the laser ablation plasma method can obtain material information without damaging the overall structure of the material, and the background interference, instrument noise and system error are effectively removed through the spectrum pre-processing process, so as to ensure the accuracy of subsequent spectral line identification and quantitative analysis, not only improve the detection efficiency, but also provide an effective means for health monitoring and fault warning of in-service equipment.

[0056] In the embodiment of the present application, the spectrum line identification of the pre-processed spectrum data is performed in step S200 to determine the ion spectral line and the atomic spectral line of the target element, including the following steps C1-C3:

[0057] C1: determining the spectral peak position according to the pre-processed spectrum data.

[0058] C2: comparing the spectral peak position with the atomic spectrum database to determine the element attribution of the spectrum data.

[0059] C3: determining the ion spectral line and the atomic spectral line of the target element according to the element attribution of the spectral data.

[0060] Specifically, in step C1, the spectral peak position is determined according to the pre-processed spectral data, and the specific operation can be:

[0061] The pre-processed spectral data is peak-searched by using a continuous wavelet transform algorithm, and the spectral peak position in the pre-processed spectral data is detected by the change of the wavelet coefficient.

[0062] The center wavelength of the spectral peak is obtained according to the spectral peak position in the pre-processed spectral data.

[0063] Specifically, in step C2, the spectral peak position is compared with the atomic spectral database to determine the element attribution of the spectral data, which means that the center wavelength of the spectral peak is compared with the standard wavelength in the atomic spectral database to determine the element attribution of the spectral data.

[0064] Specifically, in step C3, the ion spectral line and the atomic spectral line of the target element are determined according to the element attribution of the spectral data, and the specific operation can be:

[0065] The spectral data is attributed and divided according to the element attribution of the spectral data to obtain a spectral line set of the target element.

[0066] The ion spectral line and the atomic spectral line of the target element are selected according to the spectral line set of the target element.

[0067] For example, taking the target element as Ca element, the Ca element spectral line set is identified from the spectral data, and two spectral lines are selected from the Ca element spectral line set, which are the ion spectral line Ca II and the atomic spectral line Ca I of the Ca element, as the target spectral lines for subsequent spectral feature quantity calculation.

[0068] In an optional embodiment, in step S200, the spectral line identification on the pre-processed spectral data can also be performed by using a deep convolutional neural network combined with an attention mechanism, training the network to automatically learn the spectral feature pattern, using the attention mechanism to focus on the key spectral line region related to the bending intensity, realizing end-to-end automatic identification and classification of the spectral line, avoiding the limitations of the traditional method relying on prior knowledge and manually setting parameters, and improving the identification accuracy in the spectral environment.

[0069] In another optional embodiment, in step S200, the spectral line identification on the pre-processed spectral data can also be performed by using a fuzzy clustering algorithm combined with a genetic algorithm optimization, using fuzzy clustering to group and classify similar spectral lines, and using the genetic algorithm to optimize the clustering parameters and feature weights to identify the characteristic spectral lines in different material systems, which is suitable for spectral analysis of multi-element composite materials.

[0070] In the embodiments of the present application, the spectral characteristic quantity of the target element is determined according to the ion spectral line and the atomic spectral line in step S200, including the following steps D1-D2:

[0071] D1: obtaining the spectral line intensity of the ion spectral line and the spectral line intensity of the atomic spectral line according to the ion spectral line and the atomic spectral line of the target element.

[0072] D2: obtaining the spectral characteristic quantity of the target element based on the spectral line intensity of the ion spectral line and the spectral line intensity of the atomic spectral line.

[0073] Specifically, in step D1, the spectral line intensity of the ion spectral line and the spectral line intensity of the atomic spectral line are obtained, and the specific operation can be:

[0074] reading the spectral intensity value of the corresponding wavelength position according to the ion spectral line of the target element to obtain the spectral line intensity of the ion spectral line.

[0075] reading the spectral intensity value of the corresponding wavelength position according to the atomic spectral line of the target element to obtain the spectral line intensity of the atomic spectral line.

[0076] Specifically, in step D2, the spectral characteristic quantity of the target element is obtained based on the spectral line intensity of the ion spectral line and the spectral line intensity of the atomic spectral line, which means calculating the ratio of the spectral line intensity of the ion spectral line to the spectral line intensity of the atomic spectral line to obtain the spectral characteristic quantity of the target element.

[0077] For example, taking Ca element as an example, the spectral line intensity of the ion spectral line Ca II and the spectral line intensity of the atomic spectral line Ca I of the Ca element are obtained, the ratio of the two is calculated, the spectral line intensity ratio is obtained as the spectral characteristic quantity R' of the target element, which is used for subsequent bending strength calibration analysis.

[0078] It should be noted that the present application realizes automatic recognition of spectral lines and determination of element attribution through continuous wavelet transform peak finding and atomic spectral database comparison, and obtains a spectral characteristic quantity representing the ionization degree of the material through ion-atomic spectral line intensity ratio calculation. Compared with the traditional spectrum analysis which relies on manual experience recognition, the present application solves the problems of strong subjectivity and low efficiency of spectral line recognition, especially through the automatic spectral line recognition process to ensure the objectivity and reproducibility of the results. At the same time, the ion-atomic spectral line intensity ratio can reflect the ionization characteristics of the material in the laser ablation process, which is closely related to the microstructure and chemical bonding state of the material, and further related to the macro bending strength performance. Therefore, the present application not only improves the scientificity of feature extraction, but also provides a solid theoretical basis for establishing an accurate and reliable bending strength detection model.

[0079] In the embodiment of the present application, the standard spectral feature quantity is obtained based on the standard sample in step S300, and the calibration model is obtained according to the standard spectral feature quantity, including the following steps E1-E3:

[0080] E1: Obtain the spectral data of the standard sample with a plurality of known bending strength values.

[0081] E2: Determine the standard spectral feature quantity according to the spectral data of the standard sample.

[0082] E3: Perform linear regression analysis on the standard spectral feature quantity and the known bending strength value to obtain the calibration model.

[0083] Specifically, in step E1, the spectral data of the standard sample with a plurality of known bending strength values is obtained, and the specific operation can be:

[0084] Select a plurality of standard samples with known and different bending strength values to establish a standard sample library.

[0085] Select N test points for each standard sample to perform laser-induced breakdown spectroscopy test, and obtain the spectral data of each test point.

[0086] For example, take glass fiber reinforced unsaturated polyester molding plastic as an example, select glass fiber reinforced unsaturated polyester molding plastic standard samples with different known bending strength values, randomly select 10 positions on the surface of each standard sample under the same laser-induced breakdown spectroscopy equipment parameters, and test 20 times at each position and record the spectral data.

[0087] Specifically, in step E2, the standard spectral feature quantity is determined according to the spectral data of the standard sample, and the specific operation can be:

[0088] Calculate the average value of the spectral data of N test points of each standard sample to obtain the average spectral data of the standard sample.

[0089] Pretreat and identify the spectral lines of the average spectral data to obtain the standard spectral feature quantity.

[0090] For example, take glass fiber reinforced unsaturated polyester molding plastic as an example, select glass fiber reinforced unsaturated polyester molding plastic standard samples with different known bending strength values, randomly select 10 positions on the surface of each standard sample under the same laser-induced breakdown spectroscopy equipment parameters, and test 20 times at each position and record the spectral data.

[0091] Specifically, in step E3, the standard spectral feature quantity and the known bending strength value are combined to perform linear regression analysis to obtain the calibration model, and the specific operation can be:

[0092] The standard spectral characteristic quantity of each standard sample is taken as the independent variable, and the known bending strength value is taken as the dependent variable to establish a data pair set.

[0093] The data pair set is subjected to linear regression analysis by the least square method to determine the slope and intercept parameters of the linear relationship, and the calibration model is obtained according to the slope and intercept parameters.

[0094] For example, taking Ca element as an example, the spectral line intensity ratio R of the ion spectral line and the atomic spectral line of the Ca element of the glass fiber reinforced unsaturated polyester molding plastic standard sample is taken as the independent variable, and the corresponding bending strength value BS is taken as the dependent variable, and the calibration model is obtained by linear fitting by the least square method. The calibration model can be embodied by the following formula:

[0095] R = 0.01574BS-0.08037;

[0096] Wherein, R is the spectral line intensity ratio of the ion spectral line and the atomic spectral line of the Ca element; BS is the bending strength value, unit: MPa, the goodness of fit R 2 of the calibration model and the original data = 0.9755.

[0097] In an optional embodiment, the standard spectral characteristic quantity is obtained based on the standard sample in step S300, and the calibration model is obtained according to the standard spectral characteristic quantity. The nonlinear relationship between the spectral characteristic quantity and the bending strength can also be mapped to a high-dimensional space for linear regression by support vector regression combined with kernel function mapping. Different kernel functions are used to adapt to the nonlinear relationship of material performance, and model parameters are optimized by cross-validation to improve the generalization ability and prediction accuracy of the model, which is suitable for performance prediction of multi-component composite materials.

[0098] In another optional embodiment, the standard spectral characteristic quantity is obtained based on the standard sample in step S300, and the calibration model is obtained according to the standard spectral characteristic quantity. A plurality of decision tree ensemble models can also be constructed by random forest algorithm combined with feature importance evaluation to automatically select the spectral characteristic parameters that have the greatest impact on the bending strength. Through ensemble learning, the overfitting risk of a single model is reduced, and the stability and reliability of the model are improved.

[0099] It should be noted that the present application establishes a quantitative relationship model between the spectral characteristic quantity and the bending strength by establishing a plurality of standard sample libraries with known bending strength, using multi-point average measurement and linear regression analysis; compared with the traditional material performance detection which relies on empirical formula or theoretical calculation, the present application solves the problem that the theoretical model is difficult to accurately describe the material system through the experimental data driven calibration modeling method, especially through the multi-point measurement and average value taking of each standard sample, the influence of the inherent spectral fluctuation of the laser-induced breakdown spectroscopy technology on the modeling accuracy is effectively reduced, and the least square method linear regression is used to ensure the statistical optimality of the model parameters, which not only improves the accuracy of the bending strength prediction, but also provides a reproducible modeling method for the rapid detection of different material systems, and realizes the effect from qualitative analysis to quantitative detection.

[0100] In the embodiment of the present application, the step S400 of determining the bending strength value of the sample to be measured according to the spectral characteristic quantity and the calibration model comprises the following steps F1-F2:

[0101] F1: substituting the spectral characteristic quantity of the target element into the calibration model.

[0102] F2: performing parameter calculation on the calibration model based on the spectral characteristic quantity of the target element to obtain the bending strength value of the sample to be measured.

[0103] For example, taking the Ca element as an example, the step F2 obtains the bending strength value of the sample to be measured, which can be embodied by the following formula:

[0104]

[0105] Wherein, BS' is the bending strength value of the sample to be measured, and the unit is MPa; R' is the spectral characteristic quantity of the target element.

[0106] In an optional embodiment, the step S400 of determining the bending strength value of the sample to be measured according to the spectral characteristic quantity and the calibration model can also be combined with uncertainty quantification through Bayesian inference to give a prediction confidence interval while predicting the bending strength value, and through the Bayesian framework, the prior knowledge and the observation data are fused to quantify the uncertainty of the model prediction, provide risk evaluation basis for engineering decision, and be suitable for the evaluation of power equipment materials with high safety requirements.

[0107] In another optional embodiment, the step S400 of determining the bending strength value of the sample to be measured according to the spectral characteristic quantity and the calibration model can also be combined with weight self-adaptive allocation through model fusion technology to integrate the prediction models established by a plurality of different algorithms, adjust the fusion weight according to the historical prediction accuracy of each model, improve the accuracy and robustness of the final prediction result, and reduce the influence of the prediction error of a single model. In another optional embodiment, the step S400 of determining the bending strength value of the sample to be measured according to the spectral characteristic quantity and the calibration model can also be combined with weight self-adaptive allocation through model fusion technology to integrate the prediction models established by a plurality of different algorithms, adjust the fusion weight according to the historical prediction accuracy of each model, improve the accuracy and robustness of the final prediction result, and reduce the influence of the prediction error of a single model.

[0108] It should be noted that the present application realizes rapid quantitative detection of bending strength by substituting the spectral characteristic quantity of the material to be tested into the pre-established calibration model. Compared with the traditional destructive bending strength detection method, the present application solves the problems of the need to prepare standard samples, long time consumption, high cost and damage to the original component. In particular, the quasi-non-destructive detection is realized through non-contact laser ablation and spectral analysis, so that the bending strength monitoring of in-service equipment and important components becomes possible. At the same time, the subjectivity of manual judgment is avoided through quantitative calculation of the mathematical model, which not only improves the detection efficiency and convenience, but also provides an effective technical means for the state monitoring and preventive maintenance of key infrastructure such as power facilities.

[0109] In summary, the present application obtains spectral data of the sample to be tested by using laser-induced breakdown spectroscopy technology and pre-processes the spectral data, solves the problem of destructive sampling in traditional bending strength detection, adopts the method of laser ablation to generate plasma to only generate microscopic ablation pits on the surface of the material, avoids irreversible damage to the original structure, and realizes quasi-non-destructive detection. The peak-seeking algorithm of continuous wavelet transform and the atomic spectral database comparison are used to realize automatic identification of spectral lines, and the ion atomic spectral line intensity ratio is selected as the spectral characteristic quantity, which can reflect the ionization characteristics of the material in the laser ablation process, is closely related to the microstructure and chemical bonding state of the material, and establishes a scientific internal relationship with the macro bending strength performance, has stronger stability and anti-interference ability. By establishing a plurality of known bending strength standard sample libraries, using multi-point average measurement and linear regression analysis to establish a quantitative relationship model of spectral characteristic quantity and bending strength, and using an experimental data driven modeling method to effectively reduce the influence of spectral volatility on modeling accuracy, the least square linear regression analysis ensures the statistical optimality of the model parameters, and the effect from qualitative analysis to quantitative detection is realized. The continuous monitoring of the running equipment is realized, the real-time change trend of the material performance can be captured, and an effective technical means is provided for the safe operation and intelligent management of the power system, achieving the real-time, continuous and efficient detection effect that cannot be achieved by the traditional detection method.

[0110] Embodiment 3 is a third embodiment of the present application, which provides a system for rapidly detecting the bending strength of a material, comprising: a spectral data module for obtaining spectral data of a sample to be tested, pre-processing the spectral data, and obtaining pre-processed spectral data; a spectral characteristic quantity module for identifying spectral lines of the pre-processed spectral data, determining ion spectral lines and atomic spectral lines of target elements, and determining spectral characteristic quantities according to the ion spectral lines and the atomic spectral lines; a calibration model module for obtaining standard spectral characteristic quantities based on standard samples, and obtaining a calibration model according to the standard spectral characteristic quantities; and a bending strength module for determining a bending strength value of the sample to be tested according to the spectral characteristic quantities and the calibration model.

[0111] Embodiment 4, which is the fourth embodiment of the present application, differs from the first three embodiments in that, as shown in Figure 3 In light of the above, the technical solutions of the present application or the part of the technical solutions that essentially contribute to the prior art, or the part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0112] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a list of executable instructions for implementing logic functions, which can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus or device, such as a computer-based system, a system including a processor or other system that can fetch the instructions from the instruction execution system, apparatus or device and execute the instructions, or in conjunction with these instruction execution systems, apparatus or devices. For the purpose of this specification, the "computer-readable medium" can be any device that can contain, store, communicate, propagate or transport programs for use by or in connection with an instruction execution system, apparatus or device, or in conjunction with these instruction execution systems, apparatus or devices.

[0113] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electrical devices), a portable computer diskette (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be electronically obtained, for example, by optical scanning of the paper or other medium, followed by electronic editing, interpretation or processing, if necessary, in other suitable ways, and then stored in a computer memory.

[0114] It should be understood that portions of the application can be implemented in hardware, software, firmware, or combinations thereof. In the embodiments described above, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following technologies known in the art or a combination thereof can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0115] Example 5, refer to Figures 4 to 6 As a fifth embodiment of the present application, a method for rapidly detecting the bending strength of a material is provided, and scientific demonstration is carried out through experiments to verify the beneficial effects of the present application.

[0116] In this embodiment, glass fiber reinforced unsaturated polyester molding plastics are selected as experimental objects for verification experiments. Five glass fiber reinforced unsaturated polyester molding plastics standard samples with different bending strengths are selected, and the bending strengths are 85 MPa, 110 MPa, 135 MPa, 160 MPa and 185 MPa, respectively. The bending strength value of each standard sample is obtained by the traditional three-point bending test method according to the GB / T 1449-2005 standard. A laser-induced breakdown spectroscopy device is used to test each standard sample. Ten test points are randomly selected on the surface of each sample, and 20 laser ablations are performed on each test point. The spectral data are recorded and the average value is calculated to reduce the influence of accidental errors.

[0117] As shown in Figure 4 and Figure 5 The original spectrum and the spectrum after baseline correction and noise reduction are shown in the schematic diagram. As can be seen from the schematic diagram, there are obvious background noise interference, baseline drift and signal fluctuation in the original spectrum data, which seriously affect the accuracy of subsequent spectral line identification and quantitative analysis. After baseline correction and noise reduction, the spectral line characteristics are clearer, the spectral baseline is stable, the noise level is greatly reduced, the spectral line peak shape is more sharp and clear, and the signal-to-noise ratio and signal-to-background ratio are improved. It is shown that by removing the background spectrum, baseline correction and noise reduction of the spectrum, and full-spectrum area normalization processing, most of the noise is effectively eliminated, the influence of laser power fluctuation is eliminated, and the quality of the spectral signal is improved.

[0118] In addition, when identifying the spectral lines of the standard sample, the ion spectral line Ca II and the atomic spectral line Ca I of the Ca element are selected from the identification results, the spectral intensity values at the corresponding wavelength positions are read, the ratio of the two is calculated as the standard spectral characteristic quantity, and a calibration model is constructed according to the standard spectral characteristic quantity, as shown in Figure 6The figure shows the calibration curve of the intensity ratio of the ion atomic spectral line of the glass fiber reinforced unsaturated polyester molding sample and the bending strength, wherein the glass fiber reinforced unsaturated polyester molding standard samples with different bending strengths are used, and the bending strength ranges from 85 MPa to 185 MPa; it can be seen from the figure that the spectral characteristic quantity and the material bending strength present a good linear relationship, the calibration model established is R=0.01574BS-0.08037, the goodness of fit R 2 reaches 0.9755, indicating that the selected intensity ratio of the ion atomic spectral line as the spectral characteristic quantity has strong characterization ability and can accurately reflect the bending strength performance of the material; at the same time, the linear relationship of the calibration curve is good and the data points are uniformly distributed without obvious outliers, indicating that the present application has stable detection performance in different bending strength ranges and can realize accurate quantitative conversion from the spectral characteristic quantity to the bending strength value, thereby providing a reliable mathematical model basis for the rapid detection of unknown samples.

[0119] It should be noted that the above examples are only used to illustrate the technical solutions of the present application and not to limit it, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A method for rapidly detecting the bending strength of a material, characterized by: The method comprises the following steps: acquiring spectral data of a sample to be measured, preprocessing the spectral data to obtain preprocessed spectral data; performing spectral line identification on the preprocessed spectral data to determine ion spectral lines and atomic spectral lines of a target element, and determining spectral feature quantities of the target element according to the ion spectral lines and the atomic spectral lines; acquiring standard spectral feature quantities based on standard samples, and obtaining a calibration model according to the standard spectral feature quantities; determining a bending intensity value of the sample to be measured according to the spectral feature quantities and the calibration model.

2. The method of rapidly determining the flexural strength of a material of claim 1, wherein: The method for performing spectral line identification on the preprocessed spectral data to determine ion spectral lines and atomic spectral lines of a target element comprises the following steps: determining spectral peak positions according to the preprocessed spectral data; comparing the spectral peak positions with an atomic spectral database to determine element attribution of the spectral data; determining ion spectral lines and atomic spectral lines of the target element according to the element attribution of the spectral data.

3. The method of rapidly determining the flexural strength of a material of claim 2, wherein: The method for determining spectral feature quantities of the target element according to the ion spectral lines and the atomic spectral lines comprises the following steps: acquiring spectral line intensities of the ion spectral lines and spectral line intensities of the atomic spectral lines according to the ion spectral lines and the atomic spectral lines of the target element; obtaining spectral feature quantities of the target element based on the spectral line intensities of the ion spectral lines and the spectral line intensities of the atomic spectral lines.

4. The method of rapidly determining the flexural strength of a material of claim 3 wherein: The method for acquiring standard spectral feature quantities based on standard samples and obtaining a calibration model according to the standard spectral feature quantities comprises the following steps: acquiring spectral data of standard samples with known bending intensity values; determining standard spectral feature quantities according to the spectral data of the standard samples; performing linear regression analysis on the standard spectral feature quantities and the known bending intensity values to obtain the calibration model.

5. The method of rapidly determining the flexural strength of a material of claim 4 wherein: The method for acquiring spectral data of a sample to be measured comprises the following steps: generating plasma by performing laser ablation on the sample to be measured; collecting emission spectra of the plasma to obtain spectral data of the sample to be measured.

6. The method of rapidly determining the flexural strength of a material of claim 5 wherein: The method for preprocessing the spectral data to obtain preprocessed spectral data comprises the following steps: removing background spectra from the spectral data; performing baseline correction and noise reduction processing on the spectral data after removing the background spectra; performing normalization processing on the spectral data after the baseline correction and the noise reduction processing to obtain the preprocessed spectral data.

7. The method of rapidly determining the flexural strength of a material of claim 3 wherein: The method for obtaining spectral feature quantities of the target element based on the spectral line intensities of the ion spectral lines and the spectral line intensities of the atomic spectral lines comprises the following steps:

8. A system for rapid detection of the bending strength of a material, applying a method for rapid detection of the bending strength of a material according to any one of claims 1 to 7, characterized in that: calculating a ratio of the spectral line intensities of the ion spectral lines to the spectral line intensities of the atomic spectral lines to obtain the spectral feature quantities of the target element. The method comprises the following steps: a spectral data module, configured to acquire spectral data of a sample to be measured, and preprocess the spectral data to obtain preprocessed spectral data; a spectral feature quantity module, configured to perform spectral line identification on the preprocessed spectral data to determine ion spectral lines and atomic spectral lines of a target element, and determine spectral feature quantities of the target element according to the ion spectral lines and the atomic spectral lines; a calibration model module, configured to acquire standard spectral feature quantities based on standard samples, and obtain a calibration model according to the standard spectral feature quantities; a bending intensity module, configured to determine a bending intensity value of the sample to be measured according to the spectral feature quantities and the calibration model. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The computer program is executed by the processor to implement the steps of the method for quickly detecting the bending strength of the material according to any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the method for quickly detecting the bending strength of the material according to any one of claims 1-7.

Images