CN119525842A - Metal welding online detection method based on LIBS technology - Google Patents

Abstract

The invention relates to an LIBS technology, in particular to a metal welding online detection method based on the LIBS technology, which comprises the following steps of (S1) identifying the position of a welding line by utilizing an image identification technology, (S2) driving a light emitting module to move until the emergent light of the light emitting module is focused at the welding line, receiving the detection light at the welding line by a light receiving module to obtain the relative light intensity corresponding to various elements, (S3) correcting the relative light intensity of the elements to be detected, (S4) obtaining the relative content of the elements by utilizing the relative light intensity, and (S5) obtaining the content of the elements to be detected according to the relative content. The invention has the advantages of on-line, accurate result and the like, and is applied to welding of metal workpieces such as steel pipes and the like.

Description

LIBS technology-based metal welding on-line detection method

Technical Field

The invention relates to LIBS technology, in particular to a metal welding on-line detection method based on LIBS technology.

Background

LIBS technology is an emission spectrum analysis method for sample analysis by using laser as a light source to excite a sample and acquiring the spectrum of elements in the sample, and the method needs to focus laser energy on the surface of the sample after beam expansion and focusing of the laser so as to acquire a high-energy spectrum.

Currently, LIBS technology is applied in the analysis of fixed samples. If the existing LIBS technology is applied to moving samples, the following technical problems are faced:

1. in view of the fixation of the light source, it is not ensured that the outgoing light of the light source is incident at the weld.

2. The outgoing light cannot be guaranteed to be focused at the weld.

3. The light intensity correction function cannot be provided in real time, i.e., the detection accuracy cannot be ensured.

Based on the technical problems, the LIBS technology cannot be applied to steel pipe welding.

Disclosure of Invention

In order to solve the defects in the prior art, the invention provides a metal welding on-line detection method based on the LIBS technology.

The invention aims at realizing the following technical scheme:

An on-line detection method for metal welding based on LIBS technology,

The detection method comprises the following steps:

(S1) the identification unit identifies the position of the welding line by using an image identification technology;

(S2) the driving unit drives the light emitting module to move until the outgoing light of the light emitting module is focused at the weld;

The detection light at the welding seam is received by a light receiving module, and the relative light intensity corresponding to various elements is obtained;

(S3) correcting the relative light intensity of the element to be detected;

(S4) obtaining the relative content of each element by using the relative light intensity;

(S5) obtaining the content of the element to be detected according to the relative content.

Compared with the prior art, the invention has the following beneficial effects:

1. Real-time and online detection is realized;

Based on the weld joint recognition result of the first recognition unit, the driving unit is used for driving the light source of the LIBS detection unit to move, and the second recognition unit works, so that the emergent light of the light source is accurately focused at the weld joint, and real-time and online detection is realized;

2. The detection accuracy is good;

by utilizing the characteristics of rapidness and accuracy of the LIBS technology, the rapid and accurate detection of metals and nonmetal at the metal weld joint is realized;

The optimal correction mode corresponding to the material model is obtained by establishing the comparison between the model and the result, and the correction mode corresponding to the sample is found when the sample is detected, so that the accuracy of the detection result is ensured;

the accuracy of the correction mode is further improved by utilizing multiple iterations of the coefficients;

and the standard sample arranged on the bearing seat is utilized to realize real-time correction, so that the detection accuracy is further improved.

Drawings

The present disclosure will become more readily understood with reference to the accompanying drawings. It is to be understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of the present invention. In the figure:

FIG. 1 is a schematic flow chart of a metal welding on-line detection method according to the present invention;

FIG. 2 is a block diagram of a detection implementation device according to the present invention;

FIG. 3 is a background image without welded steel tubing;

FIG. 4 is an acquired image of a steel pipe;

FIG. 5 is a truncated background image;

FIG. 6 is an identified region in a truncated acquired image;

FIG. 7 is a gray scale view of a truncated background image;

FIG. 8 is a gray scale map of an identified region of an acquired image;

FIG. 9 is an identification image after background subtraction;

FIG. 10 is a binarized identification image;

FIG. 11 is an identification image after eliminating noise points;

FIG. 12 is a steel tube profile image;

FIG. 13 is a steel pipe weld image;

FIG. 14 is a schematic representation of the offset point versus fitted curve position using the present invention;

fig. 15 is a schematic diagram of the positional relationship between offset points and fitted curves in the prior art.

In the drawings, 1-steel pipe, 11-light emitting module, 12-light receiving module, 13-analysis module, 21-first recognition unit, 22-second recognition unit, 31-driving unit, 41-gas supply unit.

Detailed Description

Figures 1-15 and the following description depict alternative embodiments of the invention to teach those skilled in the art how to make and reproduce the invention. In order to teach the technical solution of the present invention, some conventional aspects have been simplified or omitted. Those skilled in the art will appreciate variations or alternatives derived from these specific embodiments that fall within the scope of the invention. Those skilled in the art will appreciate that the features described below can be combined in various ways to form multiple variations of the invention. Thus, the invention is not limited to the following alternative embodiments, but only by the claims and their equivalents.

Example 1.

The metal welding on-line detection method based on the LIBS technology in the embodiment of the invention, as shown in figure 1, comprises the following steps:

(S1) the identification unit identifies the position of the welding line by using an image identification technology.

(S2) the driving unit 31 drives the light emitting module 11 to move until the outgoing light of the light emitting module 11 is focused at the weld.

The detection light at the welding seam is received by the light receiving module, and the relative light intensity corresponding to various elements is obtained.

(S3) correcting the relative light intensity of the element to be measured.

(S4) obtaining the relative content of each element by using the relative light intensity.

(S5) obtaining the content of the element to be detected according to the relative content.

In order to accurately identify the weld position, further, the identification mode is as follows:

And respectively obtaining a background image without a welding workpiece and an acquisition image of the welding workpiece, and converting the background image and the identification area in the acquisition image into a gray level image.

And subtracting the background by using the gray value of the image to obtain the identification image.

The workpiece contour is extracted from the recognition image, and an upper boundary line is obtained in the contour.

A weld is obtained in the upper boundary line.

In order to accurately obtain the upper boundary line, further, the upper boundary line is obtained in such a manner that:

Dividing the workpiece contour image into a plurality of columns, and searching the mutation points of each column from top to bottom, wherein the upper boundary line is the connecting line of the mutation points of all columns.

In order to improve the accuracy of the correction, further, the light intensity is corrected by:

and according to the material type of the welding seam, selecting a correction mode corresponding to the interference element in the material in a database, wherein the interference element in the material type and the correction mode are in one-to-one correspondence in the database.

In order to more purposefully correct the light intensity of the sample to be tested, further, the corresponding relationship in the database is obtained in the following manner:

And detecting standard samples of various material types, and obtaining coefficients corresponding to the additive interference correction mode and the multiplicative interference correction mode of the interference element by using the obtained light intensity values and nominal content values of the measuring element and the interference element.

For interference elements of a standard sample of any material type, detecting results of the interference elements in the material type are processed by comparing an additive interference correction mode, a multiplicative interference correction mode and a mixed correction mode, a correction mode with the optimal detection result is selected to correspond to the interference elements of the material type and is stored, the additive interference correction mode is int_i=int_i⁰+Σ(f_ij·C_j, the multiplicative interference correction mode is that int_i=int_i⁰·(1+Σ(f_ij·A_j)),int_i⁰、int_i is light intensity before and after correction of an ith element to be detected, C_j is content of the jth interference element, f_ij is interference coefficient of the jth interference element to the ith element to be detected, and A_j is product of the content of the interference element and the light intensity.

The hybrid correction mode is to utilize an additive interference correction mode and then a multiplicative interference correction mode.

In order to improve the accuracy of the correction mode, further, after obtaining the coefficient once, the light intensity is obtained by using the correction mode, the coefficient is obtained again, and after a plurality of iterations, the coefficient is obtained in average a plurality of times as the coefficient in the correction mode.

In order to accurately obtain the interference coefficient, a nonlinear correction model is further established by using a correction mode, wherein the content C^ˊ_i =Σ(a_k·int_i^k),a_k of the element to be detected is a coefficient, k is a non-negative integer, and the interference coefficient of each interference element to the element to be detected is fitted by using the light intensity value of the element to be detected in the standard sample and the nominal content value of each element.

In order to obtain the percentage content of the element to be detected, in the step (S5), the calculation mode is as follows:

C_%= 100·C/ (100+ΣC_j),C_%, C are the percentage content and the relative content of the element to be measured, respectively, Σc_j is the sum of the relative contents of the other elements.

In order to implement the above detection method, further, as shown in fig. 2, the present invention implements detection by using an apparatus including:

A LIBS detection unit including a light emitting module 11, a light receiving module 12, and an analyzing module 13, the light emitting module 11 including a light source and a condensing lens. The light receiving module 12 includes a light converging device and a spectrometer. When the welding device works, emergent light emitted by the light source passes through the converging lens and then is focused on the welding workpiece, and the part of the welding workpiece is vaporized by photoablation to form transient plasma and emit light radiation. The optical radiation is converged by the optical converging device and then enters the spectrometer, and an electric signal output by a detector of the spectrometer is sent to the analysis module 13, and the analysis module 13 analyzes the electric signal by utilizing the LIBS technology.

The first recognition unit 21, the first recognition unit 21 is used for recognizing the position of the metal weld.

And a driving unit 31, wherein the driving unit 31 is used for driving the light source to move, so that the emergent light of the light source is focused at the welding seam.

In order to ensure that the outgoing light is accurately focused at the weld, the device further comprises:

a second recognition unit 22, the second recognition unit 22 is used for recognizing whether the emergent light is focused at the welding seam.

For accurate identification, further, the second identification unit 22 employs a visual identification unit or a ranging unit provided on the driving unit 31 for a distance between the light source and the weld so that the distance is equal to a focal length of the converging lens.

In order to provide the real-time correction function, the apparatus further comprises.

The bearing seat is used for bearing a plurality of standard samples, and the driving unit 31 is used for driving the light source to move, so that the emergent light of the light source is focused on the standard samples, and the LIBS detection unit is corrected in real time.

Example 2.

According to the application example of the LIBS technology-based metal welding on-line detection method in the welding of the steel pipe 1 in the embodiment 1 of the invention.

In this application example, as shown in fig. 2, in the implementation device of the detection method, the LIBS detection unit includes a light emitting module 11, a light receiving module 12, and an analyzing module 13, and the light emitting module 11 includes a laser and a converging lens, and is disposed on a driving unit 31 (the present embodiment employs a three-dimensional mechanical arm, providing vertical up-down movement, horizontal rotation). The light receiving module 12 includes a light converging device and a spectrometer in which a line detector is employed.

The first recognition unit 21 adopts a visual recognition unit for recognizing the weld positions between the steel pipes 1.

The second recognition unit 22 is a distance measuring unit, specifically a laser distance meter, and is disposed on the three-dimensional mechanical arm, so as to obtain the distance between the light source (or the converging lens) and the welding seam.

The gas supply unit 41 includes an argon gas tank and a pipe, and argon gas discharged from the pipe purges the excitation of the weld.

The bearing seat is provided with various brands of steel standard samples, and the three-dimensional mechanical arm drives the light source, so that emergent light of the laser is converged on the standard samples.

The embodiment of the invention discloses a metal welding on-line detection method based on LIBS technology, which specifically detects 316 percent of manganese content in a steel pipe 1, as shown in figure 1, and comprises the following steps:

(S0) establishing a correspondence between the interference elements and the correction patterns in the material type. The method comprises the following steps:

The nonlinear correction models are built by using (additive interference and multiplicative interference) correction modes respectively, wherein the content C^ˊ_i=Σ(a_k·int_i^k),a_k of the element to be detected is a coefficient, and k is a non-negative integer, in the embodiment, k=0, 1 and 2.

In the additive interference correction mode, int_i=int_i⁰+Σ(f_ij·C_j). In the multiplicative interference correction mode, int_i=int_i⁰·(1+Σ(f_ij·A_j)).int_i⁰、int_i is the light intensity before correction (i.e. the original light intensity) and the light intensity after correction of the ith element to be detected, C_j is the content of the jth interference element, and f_ij is the interference coefficient of the jth interference element to the ith element to be detected. In this example, nickel and chromium were selected as interfering elements for the manganese 293 channel.

And detecting standard samples of various material types to obtain the relative light intensity (the ratio of the light intensity of the interference element to the light intensity of the iron) and the relative content (the ratio of the content of the interference element to the content of the iron) of the element to be detected and the interference element in the standard samples.

。

And fitting the interference coefficients corresponding to the additive interference correction mode and the multiplicative interference correction mode by using correction models corresponding to the additive interference correction mode and the multiplicative interference correction mode respectively corresponding to each interference element.

Such as the additive interference coefficient of nickel, the multiplicative interference coefficient, the additive interference coefficient of chromium and the multiplicative interference coefficient.

For each interference element, the corrected light intensity is obtained by using an additive interference correction mode, a multiplicative interference correction mode and a mixed correction mode (the corrected light intensity is obtained by using the additive interference correction mode firstly, then the corrected light intensity is substituted into the multiplicative interference correction mode as the original light intensity to obtain the final corrected light intensity), and then the relative content of the measurement element is obtained by using the mapping relation between the relative content and the relative light intensity.

And selecting a correction mode corresponding to the relative content closest to the nominal value of the standard sample as the correction mode corresponding to the interference element in the material type.

For example, the effect of the multiplicative interference correction mode corresponding to the interference element nickel is optimal, so that the multiplicative interference correction mode corresponding to the interference element nickel is optimal, and the effect of the additive interference correction mode corresponding to the interference element chromium is optimal, so that the additive interference correction mode corresponding to the interference element chromium is optimal.

After the interference coefficient is obtained, the correction light intensity is obtained by using the correction mode, and then the correction light intensity is substituted into the correction model, and the interference coefficient is fitted again. For example, iterating 199 times, obtaining 200 groups of interference coefficients, averaging, taking the average value as the interference coefficient, and storing.

Data to be used was regressed 1 st.

。

Data to be used was regressed 2 nd time.

。

Data for use was regressed 200 th time.

。

The interference coefficient fitted out by each regression.

。

And establishing the corresponding relation between the interference elements in the material types, the correction modes and the correction coefficients, and storing the corresponding relation.

(S1) the first recognition unit 21 recognizes the position of the weld by using an image recognition technique in such a manner that:

As shown in fig. 3, a background image is obtained when there is no welded steel pipe 1, and as shown in fig. 4, an acquired image of the welded steel pipe 1 is obtained.

The background image and the identified region (including the weld region) in the captured image are taken, respectively, as shown in fig. 5 and 6.

The recognition areas in the background image and the captured image are converted into grayscale images, respectively, as shown in fig. 7 and 8.

The background is subtracted by subtracting the gray values of the images to obtain an identification image, where the same is black and the different is white, as shown in fig. 9.

The identification image is binarized as shown in fig. 10.

Noise points are eliminated using a morphological erosion dilation operation, as shown in fig. 11.

The profile of the steel pipe 1 was extracted to obtain the diameter of the steel pipe 1 as shown in fig. 12.

And extracting the outline of the steel pipe 1 from the identification image, and obtaining an upper boundary line in the outline, wherein the outline image of the steel pipe 1 is divided into a plurality of columns, and the mutation points of each column are searched from top to bottom, and the upper boundary line is the connecting line of the mutation points of all columns.

The weld was obtained in the upper boundary line by obtaining all maximum points on the upper boundary line, 20 points on the left and right, and peak heights or peak areas, where two or more pixels at the peak heights were regarded as the center of the weld, and 15 pixels on the left and right of the center of the peak were regarded as the weld. As shown in fig. 13, the red spot area is the weld position.

(S2) the driving unit 31 drives the light emitting module 11 to move until the outgoing light of the light emitting module 11 is focused at the weld. The specific method is as follows:

The first recognition unit 21 recognizes the position of the weld using an image recognition technique;

The driving unit 31 drives the light emitting module 11 to move according to the position, and at the same time, the laser range finder obtains the distance between the light source and the weld, and when the distance between the converging lens and the weld is the focal length of the converging lens, the driving unit 31 stops moving.

The outgoing light emitted by the laser is focused on the welding seam through the converging lens, the light at the welding seam is received by the light receiving module 12, is split by a spectrometer and is converted into an electric signal through the linear array detector, and the electric signal is sent to the analysis module 13.

During the excitation, the laser rangefinder outputs a distance in real time, and when the distance between the converging lens and the weld deviates from the focal length of the converging lens, the driving unit 31 drives the light source to move so that the distance between the converging lens and the weld is equal to the focal length. The argon gas supplied from the gas supply unit 41 purges the excitation of the weld.

The detection light at the weld is received by the light receiving module 12, obtaining (original) relative light intensities corresponding to the respective elements.

The light intensity data is that the relative light intensity of manganese is 0.236124120043884, the relative light intensity of chromium is 3.02214616341552, and the relative light intensity of nickel is 24.2685493177653.

And (S3) selecting a correction mode and an interference coefficient corresponding to the type of the material of the current welded steel pipe 1 according to the corresponding relation in the database, and correcting the original relative light intensity of the element to be detected. The correction light intensity of the manganese after the influence of the (additive) interference element is removed is obtained by using the additive interference correction mode, and then the correction light intensity of the manganese after the (additive) interference element is removed is obtained by substituting the correction light intensity into the multiplicative interference correction mode.

In the embodiment, the corrected light intensity of Mn is 0.253568212352195, and the corrected light intensity of Mn is searched in a database, and in the material model corresponding to the steel pipe 1, the interference element nickel corresponds to the multiplicative interference correction mode, and the interference element chromium corresponds to the additive interference correction mode.

And S4, obtaining the relative content of each element according to the relative light intensity of the element to be detected and other elements by using the mapping relation between the relative content and the relative light intensity. The relative content of manganese 1.10827089728167.

And (S5) obtaining the content of the element to be detected according to the relative content and the following formula, wherein the percentage content of manganese is 1.51348291632582%.

C_%= 100·C/ (100+ΣC_j),C_% and C are the percentage content and the relative content of the element to be detected, respectively, and ΣC_j is the sum of the relative contents of other elements in the sample.

After a period of use, the drive unit 31 drives the light source to move, in cooperation with the laser range finder, so that the distance between the converging lens and the standard sample is equal to the focal length. The outgoing light emitted by the laser is focused on the standard sample through the converging lens, and the excitation light on the standard sample is received by the light receiving module 12 and sent to the analysis module 13. The correction coefficient is obtained according to the measured value output by the analysis module 13 and the nominal value of the standard sample, and is used for detecting the welding seam. The derivation of correction coefficients is prior art in the field of analytical instrumentation.

After using the correction pattern of the present invention, as shown in fig. 14, in the fitted curve (between relative content and relative light intensity), the offset point is significantly close to the curve.

In contrast, as shown in fig. 15, in the related art, the offset point is significantly deviated from the curve.

Example 3.

According to the application example of the LIBS technology-based metal welding on-line detection method in copper pipe welding in embodiment 1 of the present invention, the difference from embodiment 2 is that:

1. The second recognition unit 22 recognizes whether the outgoing light is focused at the weld or on the standard sample by an image recognition technique using a visual recognition unit.

2. Various standard samples matched with the copper pipe are configured.

3. The three-dimensional robotic arm provides translational motion in the fore-aft, side-to-side, and up-and-down directions.

Claims (10)

Translated fromChinese

1.一种基于LIBS技术的金属焊接在线检测方法，其特征在于，所述检测方法包括以下步骤：1. A metal welding online detection method based on LIBS technology, characterized in that the detection method comprises the following steps:（S1）识别单元利用图像识别技术识别出焊缝的位置；(S1) The recognition unit recognizes the position of the weld using image recognition technology;（S2）驱动单元驱动光发射模块移动，直到光发射模块的出射光聚焦在所述焊缝处；焊缝处的检测光被光接收模块接收，获得各种元素对应的相对光强；(S2) The driving unit drives the light emitting module to move until the emitted light of the light emitting module is focused on the weld; the detection light at the weld is received by the light receiving module to obtain the relative light intensity corresponding to various elements;（S3）校正待测元素的相对光强；(S3) Correcting the relative light intensity of the element to be measured;（S4）利用相对光强获得各元素的相对含量；(S4) Obtaining the relative content of each element using relative light intensity;（S5）根据所述相对含量获得待测元素的含量。(S5) Obtaining the content of the element to be measured according to the relative content.2.根据权利要求1所述的检测方法，其特征在于，识别的方式为：2. The detection method according to claim 1, characterized in that the identification method is:分别获得无焊接工件时的背景图像和焊接工件的采集图像，并将背景图像和采集图像中的识别区域转换为灰度图；A background image without a welding workpiece and a captured image of the welding workpiece are obtained respectively, and the identification areas in the background image and the captured image are converted into grayscale images;利用图像的灰度值相减扣除背景，获得识别图像；The background is subtracted by subtracting the gray value of the image to obtain the recognition image;在识别图像中提取工件轮廓，在轮廓中获得上边界线；Extract the workpiece contour in the recognition image and obtain the upper boundary line in the contour;在上边界线中获得焊缝。Get the weld in the upper boundary line.3.根据权利要求2所述的检测方法，其特征在于，获得上边界线的方式为：3. The detection method according to claim 2, characterized in that the upper boundary line is obtained by:将工件轮廓图像划分为多列，自上而下地寻获每一列的突变点，所述上边界线是所有列的突变点的连线。The workpiece contour image is divided into multiple columns, and the mutation point of each column is found from top to bottom. The upper boundary line is the line connecting the mutation points of all columns.4.根据权利要求1所述的检测方法，其特征在于，校正所述光强的方式为：4. The detection method according to claim 1, characterized in that the light intensity is corrected in the following manner:根据所述焊缝处的材料类型，在数据库中选择与该材料中干扰元素对应的校正模式，所述数据库中，材料类型中的干扰元素对应校正模式和干扰系数。According to the material type at the weld, a correction mode corresponding to the interference element in the material is selected from a database, wherein the interference element in the material type corresponds to a correction mode and an interference coefficient.5.根据权利要求4所述的检测方法，其特征在于，所述数据库中对应关系的获得方式为：5. The detection method according to claim 4, characterized in that the corresponding relationship in the database is obtained by:检测各种材料类型的标准样品，获得测量元素及干扰元素的光强值和标称含量值，获得干扰元素的加性干扰校正模式和乘性干扰校正模式分别对应的系数；Detect standard samples of various material types, obtain the light intensity values and nominal content values of the measured elements and interfering elements, and obtain the coefficients corresponding to the additive interference correction mode and the multiplicative interference correction mode of the interfering elements;针对该标准样品的同一干扰元素，比较利用加性干扰校正模式、乘性干扰校正模式以及混合校正模式处理该干扰元素的检测结果，选择检测结果最优的校正模式与该材料类型中的干扰元素对应，并保存；For the same interfering element of the standard sample, compare the detection results of the interfering element processed by the additive interference correction mode, the multiplicative interference correction mode and the mixed correction mode, select the correction mode with the best detection result corresponding to the interfering element in the material type, and save it;所述混合校正模式是先利用加性干扰校正模式，再利用乘性干扰校正模式。The mixed correction mode first utilizes the additive interference correction mode and then utilizes the multiplicative interference correction mode.6.根据权利要求5所述的检测方法，其特征在于，在获得一次系数后，利用校正模式获得光强，再次获得系数，经过多次迭代，平均多次获得的系数，作为校正模式中的系数。6. The detection method according to claim 5 is characterized in that after obtaining the coefficient once, the light intensity is obtained using the correction mode, and the coefficient is obtained again. After multiple iterations, the coefficients obtained multiple times are averaged as the coefficients in the correction mode.7.根据权利要求5所述的检测方法，其特征在于，所述加性干扰校正模式为，int_i=int_i⁰+Σ(f_ij·C_j)，所述乘性干扰校正模式为，int_i=int_i⁰·(1+Σ(f_ij·A_j))，int_i⁰、int_i分别是第i个待测元素校正前、后光强，C_j是第j个干扰元素的含量，f_ij是第j个干扰元素对第i个待测元素的干扰系数，A_j是干扰元素的含量和光强的乘积。7. The detection method according to claim 5 is characterized in that the additive interference correction mode is, int_i =int_i⁰ +Σ(f_ij ·C_j ), and the multiplicative interference correction mode is, int_i =int_i⁰ ·(1+Σ(f_ij ·A_j )), int_i⁰ and int_i are the light intensities of the i-th element to be measured before and after correction, respectively, C_j is the content of the j-th interfering element,_fij is the interference coefficient of the j-th interfering element on the i-th element to be measured, and A_j is the product of the content of the interfering element and the light intensity.8.根据权利要求7所述的检测方法，其特征在于，利用校正模式建立非线性校正模型，利用标准样品中待测元素光强值和各元素标称含量值，拟合出不同校正模式下各干扰元素对待测元素的干扰系数。8. The detection method according to claim 7 is characterized in that a nonlinear correction model is established using a correction mode, and the interference coefficient of each interfering element on the element to be measured under different correction modes is fitted using the light intensity value of the element to be measured and the nominal content value of each element in the standard sample.9.根据权利要求8所述的检测方法，其特征在于，所述校正模型为：9. The detection method according to claim 8, characterized in that the calibration model is:待测元素含量C^ˊ_i =Σ(a_k·int_i^k)，a_k是系数，k是非负整数。The content of the element to be measured is C^ˊ_i =Σ(_ak ·int_i^k ), a_k is the coefficient, and k is a non-negative integer.10.根据权利要求1所述的检测方法，其特征在于，步骤（S5）中，计算方式为：10. The detection method according to claim 1, characterized in that in step (S5), the calculation method is:C_%= 100·C/ (100+ΣC_j)，C_%、C分别是待测元素的百分比含量和相对含量，ΣC_j是其它元素的相对含量之和。C_% = 100·C/ (100+ΣC_j ), C_% and C are the percentage content and relative content of the element to be measured, respectively, and ΣC_j is the sum of the relative contents of other elements.