SIA OpenIR  > 工业控制网络与系统研究室
图像与光谱信息融合的LIBS谱线校正方法及应用研究
Alternative TitleResearch on LIBS Spectral Correction Method and Application Based on the Fusion of Plasma Image and Spectral Information
张鹏1,2
Department工业控制网络与系统研究室
Thesis Advisor于海斌 ; 孙兰香
Keyword激光诱导击穿光谱 激光诱导等离子体成像 光谱数据预处理 谱线校正 定量分析
Pages121页
Degree Discipline控制理论与控制工程
Degree Name博士
2019-05-23
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract论文针对激光诱导击穿光谱在实际测量中特征谱线易产生抖动与失真的问题,通过对激光诱导击穿光谱的谱线强度和对应的激光诱导等离子体图像进行分析,诊断激光诱导击穿光谱特征谱线的情况,以提高特征谱线稳定性和可用性为目的,开展了对激光诱导击穿光谱的补偿和校正方法与在线检测应用的研究。论文的主要研究内容包括:激光诱导等离子体和激光诱导击穿光谱的产生机理研究,基于谱线拟合的自吸收效应影响消除方法研究,融合等离子体位置特征的谱线稳定性校正方法研究,融合等离子体形貌特征的谱线稳定性校正方法研究,以及面向在线应用的多维图像特征融合方法研究。具体的研究内容和相关的创新性研究成果如下:(1)通过对激光诱导击穿光谱特征谱线形状和强度的研究,提出了一种基于近似Voigt函数的拟合方法,对受自吸收效应影响的特征谱峰进行拟合和校正。首先通过近似函数对Voigt函数进行简化;再通过无自吸收样本确定谱峰中心和理想条件下的半峰宽;进而通过对谱峰两翼受自吸收效应影响较小的光谱进行拟合,得到更接近理论线型的谱线。并通过该方法在磷矿样品镁元素含量标定中的应用,验证了该方法对LIBS检测准确性的提升。(2)研究了等离子体形貌变化与LIBS光谱信号抖动的关系,提出了融合等离子体位置特征的谱线稳定性校正方法。通过水平方向的等离子体图像判断等离子体产生位置,建立等离子体产生位置与LIBS特征谱线变化的对应模型,补偿和校正特征谱线强度。通过搭建实验平台和对铝合金样品中的多种元素进行定量分析,实验结果表明融合等离子体位置特征的谱线稳定性校正方法可以有效的提升LIBS特征谱线强度的稳定性,在结合常用标定方法后,可以进一步提升定量分析的准确性。(3)根据工业现场设备原位、在线检测的需求,提出了融合等离子体形貌特征的谱线稳定性校正方法。通过垂直同轴光路采集等离子体图像和LIBS光谱数据,提取等离子体图像主成分建立特征谱线强度校正模型。在改进的实验平台上对钢样中的多种元素的定量分析中,使用融合等离子体形貌特征的校正方法校正后的谱线强度进行标定,标定结果的准确性和稳定性都有显著的提升。(4)根据对课题组自研的LIBS在线成分分析仪在磷矿选矿现场实验采集的等离子体图像和光谱数据的分析,提出了多路图像数据的降维融合方法,实现等离子体图像与光谱信息融合的谱线校正方法向在线应用的转移。通过水平和垂直方向的等离子体图像,诊断激光激发矿浆产生的等离子体形貌特征,分析其与LIBS特征谱线强度抖动的关系,以等离子体图像的轴向投影作为图像特征,建立谱线强度校正模型,提升在线检测的稳定性和可靠性。
Other AbstractFocusing on the fluctuation and distortion of the spectral line in the practical measurement, the laser-induced breakdown spectroscopy is developed by analyzing the intensity of the spectral lines of the laser-induced breakdown spectrum and the corresponding image of the laser-induced plasma. Then, diagnosing the characteristic spectral lines of the laser-induced breakdown spectrum is applied to improve the stability and usability of the characteristic spectral lines. Spectrum compensation and correction methods and on-line detection applications are studied. The main research contents of this paper include: generation mechanism of laser-induced plasma and laser-induced breakdown spectroscopy, self-absorption effect elimination method based on spectral line fitting, spectral line correction method based on the fusion of plasma location information, spectral line correction method based on the fusion of principal component information of plasma image, and the application of the fusion method in online analysis of the phosphate rock slurry. Specific research contents and related markable results are as follows: (1) By studying the shape and intensity of the characteristic spectral lines of laser-induced breakdown spectra, a fitting method based on approximate Voigt function is proposed to fit and correct the characteristic spectral peaks affected by self-absorption effect. Firstly, Voigt function is simplified by approximation function, then the center of peak and the half-peak width under ideal conditions are determined by non-self-absorption sample, and then the spectrum lines close to the theoretical line shape are obtained by fitting the spectrum whose flanks are less affected by self-absorption effect. Through the application of this method in the calibration of magnesium content in phosphate rock samples, the accuracy and reliability of LIBS detection are verified. (2) The relationship between the plasma morphology change and LIBS spectral signal jitter is studied, and a method of spectral line intensity correction based on plasma location information is proposed. The position of plasma is judged by plasma image in horizontal direction, and the corresponding model between the position of plasma generation and the change of LIBS characteristic line is established to compensate and correct the intensity of characteristic line. The experimental platform and the quantitative analysis of various elements in aluminium alloy samples were established. The experimental results show that the method of plasma morphology assisted analysis can effectively enhance the stability of LIBS characteristic spectral line strength, and the accuracy of quantitative analysis can further be improved after combining with common calibration methods. (3) According to the requirement of in-situ and on-line detection of industrial field equipment, an improved method of plasma image aided correction of spectral line intensity is proposed. The plasma image and LIBS spectral data were collected by the vertical coaxial optical path, and the principal components of the plasma image were extracted to establish the characteristic spectral line intensity correction model. In the quantitative analysis of various elements in steel samples on the improved experimental platform, the spectral line intensity calibrated by plasma image assisted correction has significantly improved the accuracy and stability of the calibration results. (4) Based on the plasma image and spectral data collected by the LIBS on-line component analyzer developed by our research group in the field experiment of phosphate ore dressing, an improved method of assistant correction of spectral line intensity by plasma morphological characteristics is proposed. Through plasma image in horizontal and vertical direction, the plasma morphology characteristics produced by laser-activated slurry are diagnosed, and the relationship between the plasma morphology characteristics and LIBS characteristic spectral line intensity jitter is analyzed. The stability and reliability of on-line detection are improved by stabilizing the spectral line intensity through plasma image characteristics.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/25156
Collection工业控制网络与系统研究室
Affiliation1.中国科学院沈阳自动化研究所
2.中国科学院大学
Recommended Citation
GB/T 7714
张鹏. 图像与光谱信息融合的LIBS谱线校正方法及应用研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2019.
Files in This Item:
File Name/Size DocType Version Access License
图像与光谱信息融合的LIBS谱线校正方法(7207KB)学位论文 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[张鹏]'s Articles
Baidu academic
Similar articles in Baidu academic
[张鹏]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[张鹏]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.