CN-122016673-A - Method and system for analyzing chemical components of metal sheet based on spectral analysis

CN122016673ACN 122016673 ACN122016673 ACN 122016673ACN-122016673-A

Abstract

The invention relates to the technical field of metal material detection, in particular to a method and a system for analyzing chemical components of a metal sheet based on spectrum analysis, which solve the technical problem of low measurement precision in the prior art. The method comprises the steps of respectively obtaining detection data of each acquisition point of a metal sheet during current detection and multiple historical detection, wherein the detection data comprise real-time working condition data and reference spectra corresponding to the acquisition points, determining at least one similar historical detection for the acquisition points from the multiple historical detection according to the difference degree between the real-time working condition data of the acquisition points during the current detection and the multiple historical detection for each acquisition point, correcting the reference spectra of the acquisition points during the current detection according to the reference spectra of the acquisition points during the at least one similar historical detection for each acquisition point to obtain corrected spectra of the acquisition points, and analyzing chemical components of the metal sheet according to the corrected spectra of each acquisition point.

Inventors

ZHANG YING
LI LU
GUO HONGBO

Assignees

陕西长空齿轮有限责任公司

Dates

Publication Date: 20260512
Application Date: 20260414

Claims (10)

1. A method for analyzing a chemical composition of a metal sheet based on spectroscopic analysis, comprising: Respectively acquiring detection data of each acquisition point of the metal sheet during current detection and repeated historical detection, wherein the detection data comprises real-time working condition data and a reference spectrum corresponding to the acquisition points; For each acquisition point, determining at least one similar historical detection for the acquisition point from multiple historical detections according to the degree of difference between the real-time working condition data of the acquisition point during the current detection and the multiple historical detections; correcting the reference spectrum of the acquisition point during the current detection according to the reference spectrum of the acquisition point during the detection of the at least one similar history aiming at each acquisition point to obtain a corrected spectrum of the acquisition point; the sheet metal is subjected to a chemical composition analysis based on the corrected spectrum for each of the acquisition points.
2. The spectroscopic analysis-based sheet metal chemical composition analysis method according to claim 1, wherein for each acquisition point, determining at least one similar historical detection for the acquisition point from among a plurality of historical detections based on a degree of difference between real-time operating condition data of the acquisition point at a current detection and the plurality of historical detections, comprises: For each acquisition point, calculating the working condition difference degree between the real-time working condition data of the acquisition point in the current detection and the real-time working condition data of the acquisition point in each historical detection respectively, wherein the working condition difference degree is used for representing the deviation degree between the working condition state of the acquisition point in the current detection and the working condition state in the historical detection; and screening at least one similar historical detection from the multiple historical detections according to the working condition difference degree.
3. The method of analyzing chemical components of a metal sheet based on spectroscopic analysis according to claim 2, wherein selecting at least one similar history from the plurality of history according to the degree of difference in operating conditions comprises: Ascending order is carried out on the plurality of times of history detection according to the working condition difference degree, and a history detection sequence is obtained; and taking the history detection of the preset number which is ranked first in the history detection sequence as the at least one similar history detection.
4. The method of analyzing a chemical composition of a metal sheet based on spectral analysis according to claim 1, wherein for each collection point, correcting the reference spectrum of the collection point at the current detection based on the reference spectrum of the collection point at the time of the at least one similar historical detection to obtain a corrected spectrum of the collection point, comprising: for each acquisition point, determining a spectrum correction weight corresponding to each similar history detection according to the difference between the reference spectrum of the acquisition point in the current detection and the reference spectrum of the acquisition point in each similar history detection, wherein the spectrum correction weight is used for representing the correction contribution degree of the reference spectrum corresponding to the similar history detection to the current detection; And carrying out weighted correction on the reference spectrum of the acquisition point during the current detection based on the spectrum correction weight to obtain a corrected spectrum of the acquisition point.
5. The method according to claim 4, wherein determining, for each of the collection points, a spectral correction weight corresponding to each of the similarity history detections based on a difference between a reference spectrum of the collection point at the time of the current detection and a reference spectrum of the collection point at each of the similarity history detections, respectively, comprises: For each acquisition point, determining the matching degree of the reference spectrum of the current detection and each similar history detection according to the difference between the reference spectrum of the acquisition point in the current detection and the reference spectrum of the acquisition point in each similar history detection; And determining a spectrum correction weight corresponding to each similar history detection according to the reference spectrum matching degree.
6. The method according to claim 5, wherein determining, for each of the collection points, a degree of matching between the current detection and the reference spectrum of each of the similar history detections based on a difference between the reference spectrum of the collection point at the current detection and the reference spectrum of the collection point at each of the similar history detections, respectively, comprises: for each acquisition point, calculating an initial difference value of a reference spectrum of the acquisition point in the current detection and a characteristic wavelength of the reference spectrum of the acquisition point in each similar historical detection, wherein the characteristic wavelength is used for representing the reaction wavelength of chemical components of the metal sheet in the spectrum; For each acquisition point, calculating a stability index between the reference spectrum of the acquisition point in the current detection and the reference spectrum of the acquisition point in each similar historical detection, wherein the stability index is used for representing the stability degree of the overall difference between the reference spectrum of the acquisition point in the current detection and the reference spectrum of the acquisition point in the corresponding similar historical detection; and determining the reference spectrum matching degree of the current detection and each similar historical detection according to the initial difference value and the stability index.
7. The method according to claim 4, wherein the step of performing weighted correction on the reference spectrum of the collection point at the current detection based on the spectral correction weight to obtain a corrected spectrum of the collection point comprises: According to the spectrum correction weight corresponding to each similar history detection, weighting the reference spectrum of the acquisition point during each similar history detection to obtain a spectrum correction amount corresponding to each similar history detection; And correcting the reference spectrum of the acquisition point in the current detection according to the spectrum correction amount corresponding to each similar history detection, and obtaining the corrected spectrum.
8. The spectroscopic-analysis-based metal thin plate chemical composition analysis method according to claim 1, wherein the analysis of the metal thin plate chemical composition according to the corrected spectrum of each of the collection points comprises: Extracting a plurality of spectral features from the corrected spectrum of each acquisition point to form a spectral feature vector of the acquisition point, wherein the spectral features comprise at least one of element fingerprint features, spectral morphological features, relationship ratio features and spatial neighborhood features; And inputting the spectral feature vector of each acquisition point into a chemical component prediction model to obtain a chemical component prediction value of each acquisition point.
9. The spectroscopic-analysis-based metal sheet chemical composition analysis method according to claim 8, further comprising: and carrying out data fusion by a spatial interpolation algorithm according to the chemical component predicted value of each acquisition point and the position distribution of each acquisition point on the metal sheet to obtain a chemical component two-dimensional distribution diagram of the metal sheet, wherein the chemical component two-dimensional distribution diagram is used for representing the spatial distribution state of the content of each element on the surface of the metal sheet.
10. A spectroscopic analysis-based sheet metal chemical composition analysis system, comprising: the data acquisition module is used for respectively acquiring detection data of each acquisition point of the metal sheet during current detection and repeated historical detection, wherein the detection data comprises real-time working condition data and a reference spectrum corresponding to the acquisition points; The history matching module is used for determining at least one similar history detection for each acquisition point from the plurality of times of history detection according to the matching degree between the current detection and the real-time working condition data of the acquisition point during the plurality of times of history detection; The spectrum correction module is used for correcting the reference spectrum of the acquisition point during the current detection according to the reference spectrum of the acquisition point during the detection of the at least one similar history aiming at each acquisition point to obtain a corrected spectrum of the acquisition point; And the component analysis module is used for carrying out chemical component analysis on the metal sheet according to the corrected spectrum of each acquisition point.

Description

Method and system for analyzing chemical components of metal sheet based on spectral analysis Technical Field The invention relates to the technical field of metal material detection, in particular to a method and a system for analyzing chemical components of a metal sheet based on spectrum analysis. Background With the rapid development of intelligent manufacturing technology, the metal sheet is used as a core basic raw material in the fields of machine manufacturing, automobile industry, electronic equipment and the like, and the precise control of chemical components of the metal sheet is directly related to the quality stability, process suitability and use safety of end products. In the sheet metal production line, chemical component analysis is completed rapidly and accurately, and the method is a key technical support for realizing real-time process optimization, product quality classification and intelligent sorting, and has important significance in improving production efficiency and reducing production cost. At present, in the detection method for the chemical components of the metal sheet in the industry, the traditional chemical wet analysis has complicated operation, long time consumption and destructiveness on samples, and the real-time detection requirement of the modern production line is difficult to meet. The main stream detection scheme at the present stage mostly adopts a fixed point type or line scanning spectrum technology, and chemical components are inverted by collecting spectrum signals of a metal sheet and combining a calibration model. However, the metal sheet production process is a complex system with dynamic changes, production conditions (such as plate temperature gradient, plate-shaped warping, surface oxide film thickness, running tension and the like) can fluctuate in real time, and systematic drift such as baseline deviation, characteristic peak deformation and the like of spectrum signals can be caused by the dynamic condition changes, so that a calibration model established based on standard laboratory static conditions is easy to fail, and finally measurement accuracy of chemical component analysis is insufficient and stability is poor, and high reliability requirements of intelligent manufacturing on detection data cannot be met. Disclosure of Invention In order to solve the technical problem of low measurement accuracy in the prior art, the invention aims to provide a method and a system for analyzing the chemical components of a metal sheet based on spectrum analysis, and the adopted technical scheme is as follows: The application provides a method for analyzing chemical components of a metal sheet based on spectrum analysis, which comprises the following steps: Respectively acquiring detection data of each acquisition point of the metal sheet during current detection and repeated historical detection, wherein the detection data comprises real-time working condition data and a reference spectrum corresponding to the acquisition points; For each acquisition point, determining at least one similar historical detection for the acquisition point from multiple historical detections according to the degree of difference between the real-time working condition data of the acquisition point during the current detection and the multiple historical detections; correcting the reference spectrum of the acquisition point during the current detection according to the reference spectrum of the acquisition point during the detection of the at least one similar history aiming at each acquisition point to obtain a corrected spectrum of the acquisition point; the sheet metal is subjected to a chemical composition analysis based on the corrected spectrum for each of the acquisition points. The application provides a sheet metal chemical component analysis system based on spectral analysis, which comprises: the data acquisition module is used for respectively acquiring detection data of each acquisition point of the metal sheet during current detection and repeated historical detection, wherein the detection data comprises real-time working condition data and a reference spectrum corresponding to the acquisition points; The history matching module is used for determining at least one similar history detection for each acquisition point from the plurality of times of history detection according to the matching degree between the current detection and the real-time working condition data of the acquisition point during the plurality of times of history detection; The spectrum correction module is used for correcting the reference spectrum of the acquisition point during the current detection according to the reference spectrum of the acquisition point during the detection of the at least one similar history aiming at each acquisition point to obtain a corrected spectrum of the acquisition point; And the component analysis module is used for carrying out chemical component analysis on the